├── .gitignore ├── Aptfile ├── LICENSE ├── Procfile ├── README.md ├── cfg └── yolov3.cfg ├── data └── coco.names ├── db.sqlite3 ├── detect_image ├── __init__.py ├── apps.py ├── darknet │ ├── __init__.py │ └── darknet.py ├── migrations │ └── __init__.py ├── static │ ├── css │ │ └── main.css │ └── js │ │ └── main.js ├── templates │ └── index.html ├── urls.py ├── utils │ ├── __init__.py │ └── utils.py └── views.py ├── manage.py ├── requirements.txt ├── result.png ├── runtime.txt ├── temp.png ├── website ├── __init__.py ├── settings.py ├── urls.py └── wsgi.py └── weights └── w /.gitignore: -------------------------------------------------------------------------------- 1 | bin 2 | *.pyc 3 | yolov3.weights 4 | staticfiles -------------------------------------------------------------------------------- /Aptfile: -------------------------------------------------------------------------------- 1 | libsm6 2 | libxrender1 3 | libfontconfig1 4 | libice6 -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | GNU GENERAL PUBLIC LICENSE 2 | Version 3, 29 June 2007 3 | 4 | Copyright (C) 2007 Free Software Foundation, Inc. 5 | Everyone is permitted to copy and distribute verbatim copies 6 | of this license document, but changing it is not allowed. 7 | 8 | Preamble 9 | 10 | The GNU General Public License is a free, copyleft license for 11 | software and other kinds of works. 12 | 13 | The licenses for most software and other practical works are designed 14 | to take away your freedom to share and change the works. By contrast, 15 | the GNU General Public License is intended to guarantee your freedom to 16 | share and change all versions of a program--to make sure it remains free 17 | software for all its users. We, the Free Software Foundation, use the 18 | GNU General Public License for most of our software; it applies also to 19 | any other work released this way by its authors. You can apply it to 20 | your programs, too. 21 | 22 | When we speak of free software, we are referring to freedom, not 23 | price. Our General Public Licenses are designed to make sure that you 24 | have the freedom to distribute copies of free software (and charge for 25 | them if you wish), that you receive source code or can get it if you 26 | want it, that you can change the software or use pieces of it in new 27 | free programs, and that you know you can do these things. 28 | 29 | To protect your rights, we need to prevent others from denying you 30 | these rights or asking you to surrender the rights. Therefore, you have 31 | certain responsibilities if you distribute copies of the software, or if 32 | you modify it: responsibilities to respect the freedom of others. 33 | 34 | For example, if you distribute copies of such a program, whether 35 | gratis or for a fee, you must pass on to the recipients the same 36 | freedoms that you received. You must make sure that they, too, receive 37 | or can get the source code. And you must show them these terms so they 38 | know their rights. 39 | 40 | Developers that use the GNU GPL protect your rights with two steps: 41 | (1) assert copyright on the software, and (2) offer you this License 42 | giving you legal permission to copy, distribute and/or modify it. 43 | 44 | For the developers' and authors' protection, the GPL clearly explains 45 | that there is no warranty for this free software. For both users' and 46 | authors' sake, the GPL requires that modified versions be marked as 47 | changed, so that their problems will not be attributed erroneously to 48 | authors of previous versions. 49 | 50 | Some devices are designed to deny users access to install or run 51 | modified versions of the software inside them, although the manufacturer 52 | can do so. This is fundamentally incompatible with the aim of 53 | protecting users' freedom to change the software. The systematic 54 | pattern of such abuse occurs in the area of products for individuals to 55 | use, which is precisely where it is most unacceptable. Therefore, we 56 | have designed this version of the GPL to prohibit the practice for those 57 | products. If such problems arise substantially in other domains, we 58 | stand ready to extend this provision to those domains in future versions 59 | of the GPL, as needed to protect the freedom of users. 60 | 61 | Finally, every program is threatened constantly by software patents. 62 | States should not allow patents to restrict development and use of 63 | software on general-purpose computers, but in those that do, we wish to 64 | avoid the special danger that patents applied to a free program could 65 | make it effectively proprietary. To prevent this, the GPL assures that 66 | patents cannot be used to render the program non-free. 67 | 68 | The precise terms and conditions for copying, distribution and 69 | modification follow. 70 | 71 | TERMS AND CONDITIONS 72 | 73 | 0. Definitions. 74 | 75 | "This License" refers to version 3 of the GNU General Public License. 76 | 77 | "Copyright" also means copyright-like laws that apply to other kinds of 78 | works, such as semiconductor masks. 79 | 80 | "The Program" refers to any copyrightable work licensed under this 81 | License. Each licensee is addressed as "you". "Licensees" and 82 | "recipients" may be individuals or organizations. 83 | 84 | To "modify" a work means to copy from or adapt all or part of the work 85 | in a fashion requiring copyright permission, other than the making of an 86 | exact copy. The resulting work is called a "modified version" of the 87 | earlier work or a work "based on" the earlier work. 88 | 89 | A "covered work" means either the unmodified Program or a work based 90 | on the Program. 91 | 92 | To "propagate" a work means to do anything with it that, without 93 | permission, would make you directly or secondarily liable for 94 | infringement under applicable copyright law, except executing it on a 95 | computer or modifying a private copy. Propagation includes copying, 96 | distribution (with or without modification), making available to the 97 | public, and in some countries other activities as well. 98 | 99 | To "convey" a work means any kind of propagation that enables other 100 | parties to make or receive copies. Mere interaction with a user through 101 | a computer network, with no transfer of a copy, is not conveying. 102 | 103 | An interactive user interface displays "Appropriate Legal Notices" 104 | to the extent that it includes a convenient and prominently visible 105 | feature that (1) displays an appropriate copyright notice, and (2) 106 | tells the user that there is no warranty for the work (except to the 107 | extent that warranties are provided), that licensees may convey the 108 | work under this License, and how to view a copy of this License. If 109 | the interface presents a list of user commands or options, such as a 110 | menu, a prominent item in the list meets this criterion. 111 | 112 | 1. Source Code. 113 | 114 | The "source code" for a work means the preferred form of the work 115 | for making modifications to it. "Object code" means any non-source 116 | form of a work. 117 | 118 | A "Standard Interface" means an interface that either is an official 119 | standard defined by a recognized standards body, or, in the case of 120 | interfaces specified for a particular programming language, one that 121 | is widely used among developers working in that language. 122 | 123 | The "System Libraries" of an executable work include anything, other 124 | than the work as a whole, that (a) is included in the normal form of 125 | packaging a Major Component, but which is not part of that Major 126 | Component, and (b) serves only to enable use of the work with that 127 | Major Component, or to implement a Standard Interface for which an 128 | implementation is available to the public in source code form. A 129 | "Major Component", in this context, means a major essential component 130 | (kernel, window system, and so on) of the specific operating system 131 | (if any) on which the executable work runs, or a compiler used to 132 | produce the work, or an object code interpreter used to run it. 133 | 134 | The "Corresponding Source" for a work in object code form means all 135 | the source code needed to generate, install, and (for an executable 136 | work) run the object code and to modify the work, including scripts to 137 | control those activities. However, it does not include the work's 138 | System Libraries, or general-purpose tools or generally available free 139 | programs which are used unmodified in performing those activities but 140 | which are not part of the work. For example, Corresponding Source 141 | includes interface definition files associated with source files for 142 | the work, and the source code for shared libraries and dynamically 143 | linked subprograms that the work is specifically designed to require, 144 | such as by intimate data communication or control flow between those 145 | subprograms and other parts of the work. 146 | 147 | The Corresponding Source need not include anything that users 148 | can regenerate automatically from other parts of the Corresponding 149 | Source. 150 | 151 | The Corresponding Source for a work in source code form is that 152 | same work. 153 | 154 | 2. Basic Permissions. 155 | 156 | All rights granted under this License are granted for the term of 157 | copyright on the Program, and are irrevocable provided the stated 158 | conditions are met. This License explicitly affirms your unlimited 159 | permission to run the unmodified Program. The output from running a 160 | covered work is covered by this License only if the output, given its 161 | content, constitutes a covered work. This License acknowledges your 162 | rights of fair use or other equivalent, as provided by copyright law. 163 | 164 | You may make, run and propagate covered works that you do not 165 | convey, without conditions so long as your license otherwise remains 166 | in force. You may convey covered works to others for the sole purpose 167 | of having them make modifications exclusively for you, or provide you 168 | with facilities for running those works, provided that you comply with 169 | the terms of this License in conveying all material for which you do 170 | not control copyright. Those thus making or running the covered works 171 | for you must do so exclusively on your behalf, under your direction 172 | and control, on terms that prohibit them from making any copies of 173 | your copyrighted material outside their relationship with you. 174 | 175 | Conveying under any other circumstances is permitted solely under 176 | the conditions stated below. Sublicensing is not allowed; section 10 177 | makes it unnecessary. 178 | 179 | 3. Protecting Users' Legal Rights From Anti-Circumvention Law. 180 | 181 | No covered work shall be deemed part of an effective technological 182 | measure under any applicable law fulfilling obligations under article 183 | 11 of the WIPO copyright treaty adopted on 20 December 1996, or 184 | similar laws prohibiting or restricting circumvention of such 185 | measures. 186 | 187 | When you convey a covered work, you waive any legal power to forbid 188 | circumvention of technological measures to the extent such circumvention 189 | is effected by exercising rights under this License with respect to 190 | the covered work, and you disclaim any intention to limit operation or 191 | modification of the work as a means of enforcing, against the work's 192 | users, your or third parties' legal rights to forbid circumvention of 193 | technological measures. 194 | 195 | 4. Conveying Verbatim Copies. 196 | 197 | You may convey verbatim copies of the Program's source code as you 198 | receive it, in any medium, provided that you conspicuously and 199 | appropriately publish on each copy an appropriate copyright notice; 200 | keep intact all notices stating that this License and any 201 | non-permissive terms added in accord with section 7 apply to the code; 202 | keep intact all notices of the absence of any warranty; and give all 203 | recipients a copy of this License along with the Program. 204 | 205 | You may charge any price or no price for each copy that you convey, 206 | and you may offer support or warranty protection for a fee. 207 | 208 | 5. Conveying Modified Source Versions. 209 | 210 | You may convey a work based on the Program, or the modifications to 211 | produce it from the Program, in the form of source code under the 212 | terms of section 4, provided that you also meet all of these conditions: 213 | 214 | a) The work must carry prominent notices stating that you modified 215 | it, and giving a relevant date. 216 | 217 | b) The work must carry prominent notices stating that it is 218 | released under this License and any conditions added under section 219 | 7. This requirement modifies the requirement in section 4 to 220 | "keep intact all notices". 221 | 222 | c) You must license the entire work, as a whole, under this 223 | License to anyone who comes into possession of a copy. This 224 | License will therefore apply, along with any applicable section 7 225 | additional terms, to the whole of the work, and all its parts, 226 | regardless of how they are packaged. This License gives no 227 | permission to license the work in any other way, but it does not 228 | invalidate such permission if you have separately received it. 229 | 230 | d) If the work has interactive user interfaces, each must display 231 | Appropriate Legal Notices; however, if the Program has interactive 232 | interfaces that do not display Appropriate Legal Notices, your 233 | work need not make them do so. 234 | 235 | A compilation of a covered work with other separate and independent 236 | works, which are not by their nature extensions of the covered work, 237 | and which are not combined with it such as to form a larger program, 238 | in or on a volume of a storage or distribution medium, is called an 239 | "aggregate" if the compilation and its resulting copyright are not 240 | used to limit the access or legal rights of the compilation's users 241 | beyond what the individual works permit. Inclusion of a covered work 242 | in an aggregate does not cause this License to apply to the other 243 | parts of the aggregate. 244 | 245 | 6. Conveying Non-Source Forms. 246 | 247 | You may convey a covered work in object code form under the terms 248 | of sections 4 and 5, provided that you also convey the 249 | machine-readable Corresponding Source under the terms of this License, 250 | in one of these ways: 251 | 252 | a) Convey the object code in, or embodied in, a physical product 253 | (including a physical distribution medium), accompanied by the 254 | Corresponding Source fixed on a durable physical medium 255 | customarily used for software interchange. 256 | 257 | b) Convey the object code in, or embodied in, a physical product 258 | (including a physical distribution medium), accompanied by a 259 | written offer, valid for at least three years and valid for as 260 | long as you offer spare parts or customer support for that product 261 | model, to give anyone who possesses the object code either (1) a 262 | copy of the Corresponding Source for all the software in the 263 | product that is covered by this License, on a durable physical 264 | medium customarily used for software interchange, for a price no 265 | more than your reasonable cost of physically performing this 266 | conveying of source, or (2) access to copy the 267 | Corresponding Source from a network server at no charge. 268 | 269 | c) Convey individual copies of the object code with a copy of the 270 | written offer to provide the Corresponding Source. This 271 | alternative is allowed only occasionally and noncommercially, and 272 | only if you received the object code with such an offer, in accord 273 | with subsection 6b. 274 | 275 | d) Convey the object code by offering access from a designated 276 | place (gratis or for a charge), and offer equivalent access to the 277 | Corresponding Source in the same way through the same place at no 278 | further charge. You need not require recipients to copy the 279 | Corresponding Source along with the object code. If the place to 280 | copy the object code is a network server, the Corresponding Source 281 | may be on a different server (operated by you or a third party) 282 | that supports equivalent copying facilities, provided you maintain 283 | clear directions next to the object code saying where to find the 284 | Corresponding Source. Regardless of what server hosts the 285 | Corresponding Source, you remain obligated to ensure that it is 286 | available for as long as needed to satisfy these requirements. 287 | 288 | e) Convey the object code using peer-to-peer transmission, provided 289 | you inform other peers where the object code and Corresponding 290 | Source of the work are being offered to the general public at no 291 | charge under subsection 6d. 292 | 293 | A separable portion of the object code, whose source code is excluded 294 | from the Corresponding Source as a System Library, need not be 295 | included in conveying the object code work. 296 | 297 | A "User Product" is either (1) a "consumer product", which means any 298 | tangible personal property which is normally used for personal, family, 299 | or household purposes, or (2) anything designed or sold for incorporation 300 | into a dwelling. In determining whether a product is a consumer product, 301 | doubtful cases shall be resolved in favor of coverage. For a particular 302 | product received by a particular user, "normally used" refers to a 303 | typical or common use of that class of product, regardless of the status 304 | of the particular user or of the way in which the particular user 305 | actually uses, or expects or is expected to use, the product. A product 306 | is a consumer product regardless of whether the product has substantial 307 | commercial, industrial or non-consumer uses, unless such uses represent 308 | the only significant mode of use of the product. 309 | 310 | "Installation Information" for a User Product means any methods, 311 | procedures, authorization keys, or other information required to install 312 | and execute modified versions of a covered work in that User Product from 313 | a modified version of its Corresponding Source. The information must 314 | suffice to ensure that the continued functioning of the modified object 315 | code is in no case prevented or interfered with solely because 316 | modification has been made. 317 | 318 | If you convey an object code work under this section in, or with, or 319 | specifically for use in, a User Product, and the conveying occurs as 320 | part of a transaction in which the right of possession and use of the 321 | User Product is transferred to the recipient in perpetuity or for a 322 | fixed term (regardless of how the transaction is characterized), the 323 | Corresponding Source conveyed under this section must be accompanied 324 | by the Installation Information. But this requirement does not apply 325 | if neither you nor any third party retains the ability to install 326 | modified object code on the User Product (for example, the work has 327 | been installed in ROM). 328 | 329 | The requirement to provide Installation Information does not include a 330 | requirement to continue to provide support service, warranty, or updates 331 | for a work that has been modified or installed by the recipient, or for 332 | the User Product in which it has been modified or installed. Access to a 333 | network may be denied when the modification itself materially and 334 | adversely affects the operation of the network or violates the rules and 335 | protocols for communication across the network. 336 | 337 | Corresponding Source conveyed, and Installation Information provided, 338 | in accord with this section must be in a format that is publicly 339 | documented (and with an implementation available to the public in 340 | source code form), and must require no special password or key for 341 | unpacking, reading or copying. 342 | 343 | 7. Additional Terms. 344 | 345 | "Additional permissions" are terms that supplement the terms of this 346 | License by making exceptions from one or more of its conditions. 347 | Additional permissions that are applicable to the entire Program shall 348 | be treated as though they were included in this License, to the extent 349 | that they are valid under applicable law. If additional permissions 350 | apply only to part of the Program, that part may be used separately 351 | under those permissions, but the entire Program remains governed by 352 | this License without regard to the additional permissions. 353 | 354 | When you convey a copy of a covered work, you may at your option 355 | remove any additional permissions from that copy, or from any part of 356 | it. (Additional permissions may be written to require their own 357 | removal in certain cases when you modify the work.) You may place 358 | additional permissions on material, added by you to a covered work, 359 | for which you have or can give appropriate copyright permission. 360 | 361 | Notwithstanding any other provision of this License, for material you 362 | add to a covered work, you may (if authorized by the copyright holders of 363 | that material) supplement the terms of this License with terms: 364 | 365 | a) Disclaiming warranty or limiting liability differently from the 366 | terms of sections 15 and 16 of this License; or 367 | 368 | b) Requiring preservation of specified reasonable legal notices or 369 | author attributions in that material or in the Appropriate Legal 370 | Notices displayed by works containing it; or 371 | 372 | c) Prohibiting misrepresentation of the origin of that material, or 373 | requiring that modified versions of such material be marked in 374 | reasonable ways as different from the original version; or 375 | 376 | d) Limiting the use for publicity purposes of names of licensors or 377 | authors of the material; or 378 | 379 | e) Declining to grant rights under trademark law for use of some 380 | trade names, trademarks, or service marks; or 381 | 382 | f) Requiring indemnification of licensors and authors of that 383 | material by anyone who conveys the material (or modified versions of 384 | it) with contractual assumptions of liability to the recipient, for 385 | any liability that these contractual assumptions directly impose on 386 | those licensors and authors. 387 | 388 | All other non-permissive additional terms are considered "further 389 | restrictions" within the meaning of section 10. If the Program as you 390 | received it, or any part of it, contains a notice stating that it is 391 | governed by this License along with a term that is a further 392 | restriction, you may remove that term. If a license document contains 393 | a further restriction but permits relicensing or conveying under this 394 | License, you may add to a covered work material governed by the terms 395 | of that license document, provided that the further restriction does 396 | not survive such relicensing or conveying. 397 | 398 | If you add terms to a covered work in accord with this section, you 399 | must place, in the relevant source files, a statement of the 400 | additional terms that apply to those files, or a notice indicating 401 | where to find the applicable terms. 402 | 403 | Additional terms, permissive or non-permissive, may be stated in the 404 | form of a separately written license, or stated as exceptions; 405 | the above requirements apply either way. 406 | 407 | 8. Termination. 408 | 409 | You may not propagate or modify a covered work except as expressly 410 | provided under this License. Any attempt otherwise to propagate or 411 | modify it is void, and will automatically terminate your rights under 412 | this License (including any patent licenses granted under the third 413 | paragraph of section 11). 414 | 415 | However, if you cease all violation of this License, then your 416 | license from a particular copyright holder is reinstated (a) 417 | provisionally, unless and until the copyright holder explicitly and 418 | finally terminates your license, and (b) permanently, if the copyright 419 | holder fails to notify you of the violation by some reasonable means 420 | prior to 60 days after the cessation. 421 | 422 | Moreover, your license from a particular copyright holder is 423 | reinstated permanently if the copyright holder notifies you of the 424 | violation by some reasonable means, this is the first time you have 425 | received notice of violation of this License (for any work) from that 426 | copyright holder, and you cure the violation prior to 30 days after 427 | your receipt of the notice. 428 | 429 | Termination of your rights under this section does not terminate the 430 | licenses of parties who have received copies or rights from you under 431 | this License. If your rights have been terminated and not permanently 432 | reinstated, you do not qualify to receive new licenses for the same 433 | material under section 10. 434 | 435 | 9. Acceptance Not Required for Having Copies. 436 | 437 | You are not required to accept this License in order to receive or 438 | run a copy of the Program. Ancillary propagation of a covered work 439 | occurring solely as a consequence of using peer-to-peer transmission 440 | to receive a copy likewise does not require acceptance. However, 441 | nothing other than this License grants you permission to propagate or 442 | modify any covered work. These actions infringe copyright if you do 443 | not accept this License. Therefore, by modifying or propagating a 444 | covered work, you indicate your acceptance of this License to do so. 445 | 446 | 10. Automatic Licensing of Downstream Recipients. 447 | 448 | Each time you convey a covered work, the recipient automatically 449 | receives a license from the original licensors, to run, modify and 450 | propagate that work, subject to this License. You are not responsible 451 | for enforcing compliance by third parties with this License. 452 | 453 | An "entity transaction" is a transaction transferring control of an 454 | organization, or substantially all assets of one, or subdividing an 455 | organization, or merging organizations. If propagation of a covered 456 | work results from an entity transaction, each party to that 457 | transaction who receives a copy of the work also receives whatever 458 | licenses to the work the party's predecessor in interest had or could 459 | give under the previous paragraph, plus a right to possession of the 460 | Corresponding Source of the work from the predecessor in interest, if 461 | the predecessor has it or can get it with reasonable efforts. 462 | 463 | You may not impose any further restrictions on the exercise of the 464 | rights granted or affirmed under this License. For example, you may 465 | not impose a license fee, royalty, or other charge for exercise of 466 | rights granted under this License, and you may not initiate litigation 467 | (including a cross-claim or counterclaim in a lawsuit) alleging that 468 | any patent claim is infringed by making, using, selling, offering for 469 | sale, or importing the Program or any portion of it. 470 | 471 | 11. Patents. 472 | 473 | A "contributor" is a copyright holder who authorizes use under this 474 | License of the Program or a work on which the Program is based. The 475 | work thus licensed is called the contributor's "contributor version". 476 | 477 | A contributor's "essential patent claims" are all patent claims 478 | owned or controlled by the contributor, whether already acquired or 479 | hereafter acquired, that would be infringed by some manner, permitted 480 | by this License, of making, using, or selling its contributor version, 481 | but do not include claims that would be infringed only as a 482 | consequence of further modification of the contributor version. For 483 | purposes of this definition, "control" includes the right to grant 484 | patent sublicenses in a manner consistent with the requirements of 485 | this License. 486 | 487 | Each contributor grants you a non-exclusive, worldwide, royalty-free 488 | patent license under the contributor's essential patent claims, to 489 | make, use, sell, offer for sale, import and otherwise run, modify and 490 | propagate the contents of its contributor version. 491 | 492 | In the following three paragraphs, a "patent license" is any express 493 | agreement or commitment, however denominated, not to enforce a patent 494 | (such as an express permission to practice a patent or covenant not to 495 | sue for patent infringement). To "grant" such a patent license to a 496 | party means to make such an agreement or commitment not to enforce a 497 | patent against the party. 498 | 499 | If you convey a covered work, knowingly relying on a patent license, 500 | and the Corresponding Source of the work is not available for anyone 501 | to copy, free of charge and under the terms of this License, through a 502 | publicly available network server or other readily accessible means, 503 | then you must either (1) cause the Corresponding Source to be so 504 | available, or (2) arrange to deprive yourself of the benefit of the 505 | patent license for this particular work, or (3) arrange, in a manner 506 | consistent with the requirements of this License, to extend the patent 507 | license to downstream recipients. "Knowingly relying" means you have 508 | actual knowledge that, but for the patent license, your conveying the 509 | covered work in a country, or your recipient's use of the covered work 510 | in a country, would infringe one or more identifiable patents in that 511 | country that you have reason to believe are valid. 512 | 513 | If, pursuant to or in connection with a single transaction or 514 | arrangement, you convey, or propagate by procuring conveyance of, a 515 | covered work, and grant a patent license to some of the parties 516 | receiving the covered work authorizing them to use, propagate, modify 517 | or convey a specific copy of the covered work, then the patent license 518 | you grant is automatically extended to all recipients of the covered 519 | work and works based on it. 520 | 521 | A patent license is "discriminatory" if it does not include within 522 | the scope of its coverage, prohibits the exercise of, or is 523 | conditioned on the non-exercise of one or more of the rights that are 524 | specifically granted under this License. You may not convey a covered 525 | work if you are a party to an arrangement with a third party that is 526 | in the business of distributing software, under which you make payment 527 | to the third party based on the extent of your activity of conveying 528 | the work, and under which the third party grants, to any of the 529 | parties who would receive the covered work from you, a discriminatory 530 | patent license (a) in connection with copies of the covered work 531 | conveyed by you (or copies made from those copies), or (b) primarily 532 | for and in connection with specific products or compilations that 533 | contain the covered work, unless you entered into that arrangement, 534 | or that patent license was granted, prior to 28 March 2007. 535 | 536 | Nothing in this License shall be construed as excluding or limiting 537 | any implied license or other defenses to infringement that may 538 | otherwise be available to you under applicable patent law. 539 | 540 | 12. No Surrender of Others' Freedom. 541 | 542 | If conditions are imposed on you (whether by court order, agreement or 543 | otherwise) that contradict the conditions of this License, they do not 544 | excuse you from the conditions of this License. If you cannot convey a 545 | covered work so as to satisfy simultaneously your obligations under this 546 | License and any other pertinent obligations, then as a consequence you may 547 | not convey it at all. For example, if you agree to terms that obligate you 548 | to collect a royalty for further conveying from those to whom you convey 549 | the Program, the only way you could satisfy both those terms and this 550 | License would be to refrain entirely from conveying the Program. 551 | 552 | 13. Use with the GNU Affero General Public License. 553 | 554 | Notwithstanding any other provision of this License, you have 555 | permission to link or combine any covered work with a work licensed 556 | under version 3 of the GNU Affero General Public License into a single 557 | combined work, and to convey the resulting work. The terms of this 558 | License will continue to apply to the part which is the covered work, 559 | but the special requirements of the GNU Affero General Public License, 560 | section 13, concerning interaction through a network will apply to the 561 | combination as such. 562 | 563 | 14. Revised Versions of this License. 564 | 565 | The Free Software Foundation may publish revised and/or new versions of 566 | the GNU General Public License from time to time. Such new versions will 567 | be similar in spirit to the present version, but may differ in detail to 568 | address new problems or concerns. 569 | 570 | Each version is given a distinguishing version number. If the 571 | Program specifies that a certain numbered version of the GNU General 572 | Public License "or any later version" applies to it, you have the 573 | option of following the terms and conditions either of that numbered 574 | version or of any later version published by the Free Software 575 | Foundation. If the Program does not specify a version number of the 576 | GNU General Public License, you may choose any version ever published 577 | by the Free Software Foundation. 578 | 579 | If the Program specifies that a proxy can decide which future 580 | versions of the GNU General Public License can be used, that proxy's 581 | public statement of acceptance of a version permanently authorizes you 582 | to choose that version for the Program. 583 | 584 | Later license versions may give you additional or different 585 | permissions. However, no additional obligations are imposed on any 586 | author or copyright holder as a result of your choosing to follow a 587 | later version. 588 | 589 | 15. Disclaimer of Warranty. 590 | 591 | THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY 592 | APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT 593 | HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY 594 | OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, 595 | THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 596 | PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM 597 | IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF 598 | ALL NECESSARY SERVICING, REPAIR OR CORRECTION. 599 | 600 | 16. Limitation of Liability. 601 | 602 | IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING 603 | WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS 604 | THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY 605 | GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE 606 | USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF 607 | DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD 608 | PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), 609 | EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF 610 | SUCH DAMAGES. 611 | 612 | 17. Interpretation of Sections 15 and 16. 613 | 614 | If the disclaimer of warranty and limitation of liability provided 615 | above cannot be given local legal effect according to their terms, 616 | reviewing courts shall apply local law that most closely approximates 617 | an absolute waiver of all civil liability in connection with the 618 | Program, unless a warranty or assumption of liability accompanies a 619 | copy of the Program in return for a fee. 620 | 621 | END OF TERMS AND CONDITIONS 622 | 623 | How to Apply These Terms to Your New Programs 624 | 625 | If you develop a new program, and you want it to be of the greatest 626 | possible use to the public, the best way to achieve this is to make it 627 | free software which everyone can redistribute and change under these terms. 628 | 629 | To do so, attach the following notices to the program. It is safest 630 | to attach them to the start of each source file to most effectively 631 | state the exclusion of warranty; and each file should have at least 632 | the "copyright" line and a pointer to where the full notice is found. 633 | 634 | 635 | Copyright (C) 636 | 637 | This program is free software: you can redistribute it and/or modify 638 | it under the terms of the GNU General Public License as published by 639 | the Free Software Foundation, either version 3 of the License, or 640 | (at your option) any later version. 641 | 642 | This program is distributed in the hope that it will be useful, 643 | but WITHOUT ANY WARRANTY; without even the implied warranty of 644 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 645 | GNU General Public License for more details. 646 | 647 | You should have received a copy of the GNU General Public License 648 | along with this program. If not, see . 649 | 650 | Also add information on how to contact you by electronic and paper mail. 651 | 652 | If the program does terminal interaction, make it output a short 653 | notice like this when it starts in an interactive mode: 654 | 655 | Copyright (C) 656 | This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'. 657 | This is free software, and you are welcome to redistribute it 658 | under certain conditions; type `show c' for details. 659 | 660 | The hypothetical commands `show w' and `show c' should show the appropriate 661 | parts of the General Public License. Of course, your program's commands 662 | might be different; for a GUI interface, you would use an "about box". 663 | 664 | You should also get your employer (if you work as a programmer) or school, 665 | if any, to sign a "copyright disclaimer" for the program, if necessary. 666 | For more information on this, and how to apply and follow the GNU GPL, see 667 | . 668 | 669 | The GNU General Public License does not permit incorporating your program 670 | into proprietary programs. If your program is a subroutine library, you 671 | may consider it more useful to permit linking proprietary applications with 672 | the library. If this is what you want to do, use the GNU Lesser General 673 | Public License instead of this License. But first, please read 674 | . 675 | -------------------------------------------------------------------------------- /Procfile: -------------------------------------------------------------------------------- 1 | web: gunicorn website.wsgi --log-file - -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Object-Detection-API 2 | 3 | A web-app that provides object detection using YOLOv3 and also an API. 4 | 5 | It's implemented using django framework and PyTorch (for YOLO model). The app uses cloudinary API for image hosting.
6 | 7 | ![Maintenance](https://img.shields.io/badge/Maintained%3F-Yes-green.svg) ![GitHub pull-requests](https://img.shields.io/github/issues-pr/atharva-18/Object-Detection-API.svg) ![GitHub issues](https://img.shields.io/github/issues/atharva-18/Object-Detection-API.svg)
8 | ![GitHub contributors](https://img.shields.io/github/contributors/atharva-18/Object-Detection-API.svg) ![Generic badge](https://img.shields.io/badge/Python-3.7.3-Blue.svg) 9 | 10 | ### Dependencies 11 | 12 | 26 | 27 | You also need to download the `yolo.weights` file and place it in the "weights" directory. 28 | 29 | You can download the weights by - 30 | ``` 31 | $ wget https://pjreddie.com/media/files/yolov3.weights 32 | ``` 33 | ### Usage 34 | 35 | Firstly, you need to put your cloudinary credentials (cloud_name, api_key and api_secret) in ```settings.py``` file. 36 | 37 | To run the server 38 | ``` 39 | $ pip3 install -r requirements.txt 40 | $ python3 manage.py collectstatic 41 | $ python3 manage.py runserver 42 | ``` 43 | 44 | The website is hosted at - https://object-detection-yolo.herokuapp.com 45 | 46 | The detection service is currently not working because of memory limits of Heroku. 47 | 48 | ### Web API 49 | 50 | To use the web API, you can send a POST request to - https://object-detection-yolo.herokuapp.com/detect/api 51 | 52 | #### Input 53 | 54 | You can send either of the following parameters - 55 | 56 | Parameter | Type | Description 57 | --------- | ------------------------------ | --------------------------------------------------------------------------------- 58 | image | file | Image file that you want to detect. 59 | image64 | text | Image in base64 form that you want to detect. Currently supports JPEG images only 60 | 61 | #### Result 62 | 63 | Parameter | Type | Description 64 | --------- | ------------------- | ----------------------------------------------- 65 | success | bool | Whether classification was successful or not 66 | detect | class label, float | pair of label and its confidence 67 | url | image URL | Link to the result image uploaded to cloudinary 68 | 69 | Example: {"success": true, "detect": { "dog": 0.9989, "truck": 0.9999 }, 'url':"https://example.com/image.png"}
70 | "detect" will be empty if no objects are detected. 71 | 72 | ### Example 73 | 74 | The website also shows the detection output with bounding boxes around the detected objects. There will be no box if the input doesn't contain any object. 75 | 76 | #### Input 77 | 78 | ![Dog](temp.png) 79 | 80 | #### Output 81 | 82 | ![Output](result.png) 83 | 84 | No. of objects detected - 3
85 | 86 | The result image is generated using matplotlib. 87 | 88 | ### TO-DO 89 | 90 |
    91 |
  • Move app to Google Cloud Platform
    • 92 |
  • Create a docker image
    • 93 |
94 | 95 | ### Contribute 96 | 97 | If you want to contribute and/or find any bug, feel free to do a pull request! 98 | -------------------------------------------------------------------------------- /cfg/yolov3.cfg: -------------------------------------------------------------------------------- 1 | [net] 2 | # Testing 3 | batch=1 4 | subdivisions=1 5 | # Training 6 | # batch=64 7 | # subdivisions=16 8 | width=416 9 | height=416 10 | channels=3 11 | momentum=0.9 12 | decay=0.0005 13 | angle=0 14 | saturation = 1.5 15 | exposure = 1.5 16 | hue=.1 17 | 18 | learning_rate=0.001 19 | burn_in=1000 20 | max_batches = 500200 21 | policy=steps 22 | steps=400000,450000 23 | scales=.1,.1 24 | 25 | [convolutional] 26 | batch_normalize=1 27 | filters=32 28 | size=3 29 | stride=1 30 | pad=1 31 | activation=leaky 32 | 33 | # Downsample 34 | 35 | [convolutional] 36 | batch_normalize=1 37 | filters=64 38 | size=3 39 | stride=2 40 | pad=1 41 | activation=leaky 42 | 43 | [convolutional] 44 | batch_normalize=1 45 | filters=32 46 | size=1 47 | stride=1 48 | pad=1 49 | activation=leaky 50 | 51 | [convolutional] 52 | batch_normalize=1 53 | filters=64 54 | size=3 55 | stride=1 56 | pad=1 57 | activation=leaky 58 | 59 | [shortcut] 60 | from=-3 61 | activation=linear 62 | 63 | # Downsample 64 | 65 | [convolutional] 66 | batch_normalize=1 67 | filters=128 68 | size=3 69 | stride=2 70 | pad=1 71 | activation=leaky 72 | 73 | [convolutional] 74 | batch_normalize=1 75 | filters=64 76 | size=1 77 | stride=1 78 | pad=1 79 | activation=leaky 80 | 81 | [convolutional] 82 | batch_normalize=1 83 | filters=128 84 | size=3 85 | stride=1 86 | pad=1 87 | activation=leaky 88 | 89 | [shortcut] 90 | from=-3 91 | activation=linear 92 | 93 | [convolutional] 94 | batch_normalize=1 95 | filters=64 96 | size=1 97 | stride=1 98 | pad=1 99 | activation=leaky 100 | 101 | [convolutional] 102 | batch_normalize=1 103 | filters=128 104 | size=3 105 | stride=1 106 | pad=1 107 | activation=leaky 108 | 109 | [shortcut] 110 | from=-3 111 | activation=linear 112 | 113 | # Downsample 114 | 115 | [convolutional] 116 | batch_normalize=1 117 | filters=256 118 | size=3 119 | stride=2 120 | pad=1 121 | activation=leaky 122 | 123 | [convolutional] 124 | batch_normalize=1 125 | filters=128 126 | size=1 127 | stride=1 128 | pad=1 129 | activation=leaky 130 | 131 | [convolutional] 132 | batch_normalize=1 133 | filters=256 134 | size=3 135 | stride=1 136 | pad=1 137 | activation=leaky 138 | 139 | [shortcut] 140 | from=-3 141 | activation=linear 142 | 143 | [convolutional] 144 | batch_normalize=1 145 | filters=128 146 | size=1 147 | stride=1 148 | pad=1 149 | activation=leaky 150 | 151 | [convolutional] 152 | batch_normalize=1 153 | filters=256 154 | size=3 155 | stride=1 156 | pad=1 157 | activation=leaky 158 | 159 | [shortcut] 160 | from=-3 161 | activation=linear 162 | 163 | [convolutional] 164 | batch_normalize=1 165 | filters=128 166 | size=1 167 | stride=1 168 | pad=1 169 | activation=leaky 170 | 171 | [convolutional] 172 | batch_normalize=1 173 | filters=256 174 | size=3 175 | stride=1 176 | pad=1 177 | activation=leaky 178 | 179 | [shortcut] 180 | from=-3 181 | activation=linear 182 | 183 | [convolutional] 184 | batch_normalize=1 185 | filters=128 186 | size=1 187 | stride=1 188 | pad=1 189 | activation=leaky 190 | 191 | [convolutional] 192 | batch_normalize=1 193 | filters=256 194 | size=3 195 | stride=1 196 | pad=1 197 | activation=leaky 198 | 199 | [shortcut] 200 | from=-3 201 | activation=linear 202 | 203 | 204 | [convolutional] 205 | batch_normalize=1 206 | filters=128 207 | size=1 208 | stride=1 209 | pad=1 210 | activation=leaky 211 | 212 | [convolutional] 213 | batch_normalize=1 214 | filters=256 215 | size=3 216 | stride=1 217 | pad=1 218 | activation=leaky 219 | 220 | [shortcut] 221 | from=-3 222 | activation=linear 223 | 224 | [convolutional] 225 | batch_normalize=1 226 | filters=128 227 | size=1 228 | stride=1 229 | pad=1 230 | activation=leaky 231 | 232 | [convolutional] 233 | batch_normalize=1 234 | filters=256 235 | size=3 236 | stride=1 237 | pad=1 238 | activation=leaky 239 | 240 | [shortcut] 241 | from=-3 242 | activation=linear 243 | 244 | [convolutional] 245 | batch_normalize=1 246 | filters=128 247 | size=1 248 | stride=1 249 | pad=1 250 | activation=leaky 251 | 252 | [convolutional] 253 | batch_normalize=1 254 | filters=256 255 | size=3 256 | stride=1 257 | pad=1 258 | activation=leaky 259 | 260 | [shortcut] 261 | from=-3 262 | activation=linear 263 | 264 | [convolutional] 265 | batch_normalize=1 266 | filters=128 267 | size=1 268 | stride=1 269 | pad=1 270 | activation=leaky 271 | 272 | [convolutional] 273 | batch_normalize=1 274 | filters=256 275 | size=3 276 | stride=1 277 | pad=1 278 | activation=leaky 279 | 280 | [shortcut] 281 | from=-3 282 | activation=linear 283 | 284 | # Downsample 285 | 286 | [convolutional] 287 | batch_normalize=1 288 | filters=512 289 | size=3 290 | stride=2 291 | pad=1 292 | activation=leaky 293 | 294 | [convolutional] 295 | batch_normalize=1 296 | filters=256 297 | size=1 298 | stride=1 299 | pad=1 300 | activation=leaky 301 | 302 | [convolutional] 303 | batch_normalize=1 304 | filters=512 305 | size=3 306 | stride=1 307 | pad=1 308 | activation=leaky 309 | 310 | [shortcut] 311 | from=-3 312 | activation=linear 313 | 314 | 315 | [convolutional] 316 | batch_normalize=1 317 | filters=256 318 | size=1 319 | stride=1 320 | pad=1 321 | activation=leaky 322 | 323 | [convolutional] 324 | batch_normalize=1 325 | filters=512 326 | size=3 327 | stride=1 328 | pad=1 329 | activation=leaky 330 | 331 | [shortcut] 332 | from=-3 333 | activation=linear 334 | 335 | 336 | [convolutional] 337 | batch_normalize=1 338 | filters=256 339 | size=1 340 | stride=1 341 | pad=1 342 | activation=leaky 343 | 344 | [convolutional] 345 | batch_normalize=1 346 | filters=512 347 | size=3 348 | stride=1 349 | pad=1 350 | activation=leaky 351 | 352 | [shortcut] 353 | from=-3 354 | activation=linear 355 | 356 | 357 | [convolutional] 358 | batch_normalize=1 359 | filters=256 360 | size=1 361 | stride=1 362 | pad=1 363 | activation=leaky 364 | 365 | [convolutional] 366 | batch_normalize=1 367 | filters=512 368 | size=3 369 | stride=1 370 | pad=1 371 | activation=leaky 372 | 373 | [shortcut] 374 | from=-3 375 | activation=linear 376 | 377 | [convolutional] 378 | batch_normalize=1 379 | filters=256 380 | size=1 381 | stride=1 382 | pad=1 383 | activation=leaky 384 | 385 | [convolutional] 386 | batch_normalize=1 387 | filters=512 388 | size=3 389 | stride=1 390 | pad=1 391 | activation=leaky 392 | 393 | [shortcut] 394 | from=-3 395 | activation=linear 396 | 397 | 398 | [convolutional] 399 | batch_normalize=1 400 | filters=256 401 | size=1 402 | stride=1 403 | pad=1 404 | activation=leaky 405 | 406 | [convolutional] 407 | batch_normalize=1 408 | filters=512 409 | size=3 410 | stride=1 411 | pad=1 412 | activation=leaky 413 | 414 | [shortcut] 415 | from=-3 416 | activation=linear 417 | 418 | 419 | [convolutional] 420 | batch_normalize=1 421 | filters=256 422 | size=1 423 | stride=1 424 | pad=1 425 | activation=leaky 426 | 427 | [convolutional] 428 | batch_normalize=1 429 | filters=512 430 | size=3 431 | stride=1 432 | pad=1 433 | activation=leaky 434 | 435 | [shortcut] 436 | from=-3 437 | activation=linear 438 | 439 | [convolutional] 440 | batch_normalize=1 441 | filters=256 442 | size=1 443 | stride=1 444 | pad=1 445 | activation=leaky 446 | 447 | [convolutional] 448 | batch_normalize=1 449 | filters=512 450 | size=3 451 | stride=1 452 | pad=1 453 | activation=leaky 454 | 455 | [shortcut] 456 | from=-3 457 | activation=linear 458 | 459 | # Downsample 460 | 461 | [convolutional] 462 | batch_normalize=1 463 | filters=1024 464 | size=3 465 | stride=2 466 | pad=1 467 | activation=leaky 468 | 469 | [convolutional] 470 | batch_normalize=1 471 | filters=512 472 | size=1 473 | stride=1 474 | pad=1 475 | activation=leaky 476 | 477 | [convolutional] 478 | batch_normalize=1 479 | filters=1024 480 | size=3 481 | stride=1 482 | pad=1 483 | activation=leaky 484 | 485 | [shortcut] 486 | from=-3 487 | activation=linear 488 | 489 | [convolutional] 490 | batch_normalize=1 491 | filters=512 492 | size=1 493 | stride=1 494 | pad=1 495 | activation=leaky 496 | 497 | [convolutional] 498 | batch_normalize=1 499 | filters=1024 500 | size=3 501 | stride=1 502 | pad=1 503 | activation=leaky 504 | 505 | [shortcut] 506 | from=-3 507 | activation=linear 508 | 509 | [convolutional] 510 | batch_normalize=1 511 | filters=512 512 | size=1 513 | stride=1 514 | pad=1 515 | activation=leaky 516 | 517 | [convolutional] 518 | batch_normalize=1 519 | filters=1024 520 | size=3 521 | stride=1 522 | pad=1 523 | activation=leaky 524 | 525 | [shortcut] 526 | from=-3 527 | activation=linear 528 | 529 | [convolutional] 530 | batch_normalize=1 531 | filters=512 532 | size=1 533 | stride=1 534 | pad=1 535 | activation=leaky 536 | 537 | [convolutional] 538 | batch_normalize=1 539 | filters=1024 540 | size=3 541 | stride=1 542 | pad=1 543 | activation=leaky 544 | 545 | [shortcut] 546 | from=-3 547 | activation=linear 548 | 549 | ###################### 550 | 551 | [convolutional] 552 | batch_normalize=1 553 | filters=512 554 | size=1 555 | stride=1 556 | pad=1 557 | activation=leaky 558 | 559 | [convolutional] 560 | batch_normalize=1 561 | size=3 562 | stride=1 563 | pad=1 564 | filters=1024 565 | activation=leaky 566 | 567 | [convolutional] 568 | batch_normalize=1 569 | filters=512 570 | size=1 571 | stride=1 572 | pad=1 573 | activation=leaky 574 | 575 | [convolutional] 576 | batch_normalize=1 577 | size=3 578 | stride=1 579 | pad=1 580 | filters=1024 581 | activation=leaky 582 | 583 | [convolutional] 584 | batch_normalize=1 585 | filters=512 586 | size=1 587 | stride=1 588 | pad=1 589 | activation=leaky 590 | 591 | [convolutional] 592 | batch_normalize=1 593 | size=3 594 | stride=1 595 | pad=1 596 | filters=1024 597 | activation=leaky 598 | 599 | [convolutional] 600 | size=1 601 | stride=1 602 | pad=1 603 | filters=255 604 | activation=linear 605 | 606 | 607 | [yolo] 608 | mask = 6,7,8 609 | anchors = 10,13, 16,30, 33,23, 30,61, 62,45, 59,119, 116,90, 156,198, 373,326 610 | classes=80 611 | num=9 612 | jitter=.3 613 | ignore_thresh = .5 614 | truth_thresh = 1 615 | random=1 616 | 617 | 618 | [route] 619 | layers = -4 620 | 621 | [convolutional] 622 | batch_normalize=1 623 | filters=256 624 | size=1 625 | stride=1 626 | pad=1 627 | activation=leaky 628 | 629 | [upsample] 630 | stride=2 631 | 632 | [route] 633 | layers = -1, 61 634 | 635 | 636 | 637 | [convolutional] 638 | batch_normalize=1 639 | filters=256 640 | size=1 641 | stride=1 642 | pad=1 643 | activation=leaky 644 | 645 | [convolutional] 646 | batch_normalize=1 647 | size=3 648 | stride=1 649 | pad=1 650 | filters=512 651 | activation=leaky 652 | 653 | [convolutional] 654 | batch_normalize=1 655 | filters=256 656 | size=1 657 | stride=1 658 | pad=1 659 | activation=leaky 660 | 661 | [convolutional] 662 | batch_normalize=1 663 | size=3 664 | stride=1 665 | pad=1 666 | filters=512 667 | activation=leaky 668 | 669 | [convolutional] 670 | batch_normalize=1 671 | filters=256 672 | size=1 673 | stride=1 674 | pad=1 675 | activation=leaky 676 | 677 | [convolutional] 678 | batch_normalize=1 679 | size=3 680 | stride=1 681 | pad=1 682 | filters=512 683 | activation=leaky 684 | 685 | [convolutional] 686 | size=1 687 | stride=1 688 | pad=1 689 | filters=255 690 | activation=linear 691 | 692 | 693 | [yolo] 694 | mask = 3,4,5 695 | anchors = 10,13, 16,30, 33,23, 30,61, 62,45, 59,119, 116,90, 156,198, 373,326 696 | classes=80 697 | num=9 698 | jitter=.3 699 | ignore_thresh = .5 700 | truth_thresh = 1 701 | random=1 702 | 703 | 704 | 705 | [route] 706 | layers = -4 707 | 708 | [convolutional] 709 | batch_normalize=1 710 | filters=128 711 | size=1 712 | stride=1 713 | pad=1 714 | activation=leaky 715 | 716 | [upsample] 717 | stride=2 718 | 719 | [route] 720 | layers = -1, 36 721 | 722 | 723 | 724 | [convolutional] 725 | batch_normalize=1 726 | filters=128 727 | size=1 728 | stride=1 729 | pad=1 730 | activation=leaky 731 | 732 | [convolutional] 733 | batch_normalize=1 734 | size=3 735 | stride=1 736 | pad=1 737 | filters=256 738 | activation=leaky 739 | 740 | [convolutional] 741 | batch_normalize=1 742 | filters=128 743 | size=1 744 | stride=1 745 | pad=1 746 | activation=leaky 747 | 748 | [convolutional] 749 | batch_normalize=1 750 | size=3 751 | stride=1 752 | pad=1 753 | filters=256 754 | activation=leaky 755 | 756 | [convolutional] 757 | batch_normalize=1 758 | filters=128 759 | size=1 760 | stride=1 761 | pad=1 762 | activation=leaky 763 | 764 | [convolutional] 765 | batch_normalize=1 766 | size=3 767 | stride=1 768 | pad=1 769 | filters=256 770 | activation=leaky 771 | 772 | [convolutional] 773 | size=1 774 | stride=1 775 | pad=1 776 | filters=255 777 | activation=linear 778 | 779 | 780 | [yolo] 781 | mask = 0,1,2 782 | anchors = 10,13, 16,30, 33,23, 30,61, 62,45, 59,119, 116,90, 156,198, 373,326 783 | classes=80 784 | num=9 785 | jitter=.3 786 | ignore_thresh = .5 787 | truth_thresh = 1 788 | random=1 789 | 790 | -------------------------------------------------------------------------------- /data/coco.names: -------------------------------------------------------------------------------- 1 | person 2 | bicycle 3 | car 4 | motorbike 5 | aeroplane 6 | bus 7 | train 8 | truck 9 | boat 10 | traffic light 11 | fire hydrant 12 | stop sign 13 | parking meter 14 | bench 15 | bird 16 | cat 17 | dog 18 | horse 19 | sheep 20 | cow 21 | elephant 22 | bear 23 | zebra 24 | giraffe 25 | backpack 26 | umbrella 27 | handbag 28 | tie 29 | suitcase 30 | frisbee 31 | skis 32 | snowboard 33 | sports ball 34 | kite 35 | baseball bat 36 | baseball glove 37 | skateboard 38 | surfboard 39 | tennis racket 40 | bottle 41 | wine glass 42 | cup 43 | fork 44 | knife 45 | spoon 46 | bowl 47 | banana 48 | apple 49 | sandwich 50 | orange 51 | broccoli 52 | carrot 53 | hot dog 54 | pizza 55 | donut 56 | cake 57 | chair 58 | sofa 59 | pottedplant 60 | bed 61 | diningtable 62 | toilet 63 | tvmonitor 64 | laptop 65 | mouse 66 | remote 67 | keyboard 68 | cell phone 69 | microwave 70 | oven 71 | toaster 72 | sink 73 | refrigerator 74 | book 75 | clock 76 | vase 77 | scissors 78 | teddy bear 79 | hair drier 80 | toothbrush 81 | -------------------------------------------------------------------------------- /db.sqlite3: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/atharva-18/Object-Detection-API/6286a3f0477b18da4ed0704be9b224b8210470e7/db.sqlite3 -------------------------------------------------------------------------------- /detect_image/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/atharva-18/Object-Detection-API/6286a3f0477b18da4ed0704be9b224b8210470e7/detect_image/__init__.py -------------------------------------------------------------------------------- /detect_image/apps.py: -------------------------------------------------------------------------------- 1 | from django.apps import AppConfig 2 | 3 | 4 | class DetectImageConfig(AppConfig): 5 | name = 'detect_image' 6 | -------------------------------------------------------------------------------- /detect_image/darknet/__init__.py: -------------------------------------------------------------------------------- 1 | from .darknet import Darknet -------------------------------------------------------------------------------- /detect_image/darknet/darknet.py: -------------------------------------------------------------------------------- 1 | import torch 2 | import torch.nn as nn 3 | import numpy as np 4 | 5 | 6 | class YoloLayer(nn.Module): 7 | def __init__(self, anchor_mask=[], num_classes=0, anchors=[], num_anchors=1): 8 | super(YoloLayer, self).__init__() 9 | self.anchor_mask = anchor_mask 10 | self.num_classes = num_classes 11 | self.anchors = anchors 12 | self.num_anchors = num_anchors 13 | self.anchor_step = len(anchors)/num_anchors 14 | self.coord_scale = 1 15 | self.noobject_scale = 1 16 | self.object_scale = 5 17 | self.class_scale = 1 18 | self.thresh = 0.6 19 | self.stride = 32 20 | self.seen = 0 21 | 22 | def forward(self, output, nms_thresh): 23 | self.thresh = nms_thresh 24 | masked_anchors = [] 25 | 26 | for m in self.anchor_mask: 27 | masked_anchors += self.anchors[m*self.anchor_step:(m+1)*self.anchor_step] 28 | 29 | masked_anchors = [anchor/self.stride for anchor in masked_anchors] 30 | boxes = get_region_boxes(output.data, self.thresh, self.num_classes, masked_anchors, len(self.anchor_mask)) 31 | 32 | return boxes 33 | 34 | 35 | class Upsample(nn.Module): 36 | def __init__(self, stride=2): 37 | super(Upsample, self).__init__() 38 | self.stride = stride 39 | def forward(self, x): 40 | stride = self.stride 41 | assert(x.data.dim() == 4) 42 | B = x.data.size(0) 43 | C = x.data.size(1) 44 | H = x.data.size(2) 45 | W = x.data.size(3) 46 | ws = stride 47 | hs = stride 48 | x = x.view(B, C, H, 1, W, 1).expand(B, C, H, stride, W, stride).contiguous().view(B, C, H*stride, W*stride) 49 | return x 50 | 51 | 52 | #for route and shortcut 53 | class EmptyModule(nn.Module): 54 | def __init__(self): 55 | super(EmptyModule, self).__init__() 56 | 57 | def forward(self, x): 58 | return x 59 | 60 | # support route shortcut 61 | class Darknet(nn.Module): 62 | def __init__(self, cfgfile): 63 | super(Darknet, self).__init__() 64 | self.blocks = parse_cfg(cfgfile) 65 | self.models = self.create_network(self.blocks) # merge conv, bn,leaky 66 | self.loss = self.models[len(self.models)-1] 67 | 68 | self.width = int(self.blocks[0]['width']) 69 | self.height = int(self.blocks[0]['height']) 70 | 71 | self.header = torch.IntTensor([0,0,0,0]) 72 | self.seen = 0 73 | 74 | def forward(self, x, nms_thresh): 75 | ind = -2 76 | self.loss = None 77 | outputs = dict() 78 | out_boxes = [] 79 | 80 | for block in self.blocks: 81 | ind = ind + 1 82 | if block['type'] == 'net': 83 | continue 84 | elif block['type'] in ['convolutional', 'upsample']: 85 | x = self.models[ind](x) 86 | outputs[ind] = x 87 | elif block['type'] == 'route': 88 | layers = block['layers'].split(',') 89 | layers = [int(i) if int(i) > 0 else int(i)+ind for i in layers] 90 | if len(layers) == 1: 91 | x = outputs[layers[0]] 92 | outputs[ind] = x 93 | elif len(layers) == 2: 94 | x1 = outputs[layers[0]] 95 | x2 = outputs[layers[1]] 96 | x = torch.cat((x1,x2),1) 97 | outputs[ind] = x 98 | elif block['type'] == 'shortcut': 99 | from_layer = int(block['from']) 100 | activation = block['activation'] 101 | from_layer = from_layer if from_layer > 0 else from_layer + ind 102 | x1 = outputs[from_layer] 103 | x2 = outputs[ind-1] 104 | x = x1 + x2 105 | outputs[ind] = x 106 | elif block['type'] == 'yolo': 107 | boxes = self.models[ind](x, nms_thresh) 108 | out_boxes.append(boxes) 109 | else: 110 | print('unknown type %s' % (block['type'])) 111 | 112 | return out_boxes 113 | 114 | 115 | def print_network(self): 116 | print_cfg(self.blocks) 117 | 118 | def create_network(self, blocks): 119 | models = nn.ModuleList() 120 | 121 | prev_filters = 3 122 | out_filters =[] 123 | prev_stride = 1 124 | out_strides = [] 125 | conv_id = 0 126 | for block in blocks: 127 | if block['type'] == 'net': 128 | prev_filters = int(block['channels']) 129 | continue 130 | elif block['type'] == 'convolutional': 131 | conv_id = conv_id + 1 132 | batch_normalize = int(block['batch_normalize']) 133 | filters = int(block['filters']) 134 | kernel_size = int(block['size']) 135 | stride = int(block['stride']) 136 | is_pad = int(block['pad']) 137 | pad = (kernel_size-1)//2 if is_pad else 0 138 | activation = block['activation'] 139 | model = nn.Sequential() 140 | if batch_normalize: 141 | model.add_module('conv{0}'.format(conv_id), nn.Conv2d(prev_filters, filters, kernel_size, stride, pad, bias=False)) 142 | model.add_module('bn{0}'.format(conv_id), nn.BatchNorm2d(filters)) 143 | else: 144 | model.add_module('conv{0}'.format(conv_id), nn.Conv2d(prev_filters, filters, kernel_size, stride, pad)) 145 | if activation == 'leaky': 146 | model.add_module('leaky{0}'.format(conv_id), nn.LeakyReLU(0.1, inplace=True)) 147 | prev_filters = filters 148 | out_filters.append(prev_filters) 149 | prev_stride = stride * prev_stride 150 | out_strides.append(prev_stride) 151 | models.append(model) 152 | elif block['type'] == 'upsample': 153 | stride = int(block['stride']) 154 | out_filters.append(prev_filters) 155 | prev_stride = prev_stride // stride 156 | out_strides.append(prev_stride) 157 | models.append(Upsample(stride)) 158 | elif block['type'] == 'route': 159 | layers = block['layers'].split(',') 160 | ind = len(models) 161 | layers = [int(i) if int(i) > 0 else int(i)+ind for i in layers] 162 | if len(layers) == 1: 163 | prev_filters = out_filters[layers[0]] 164 | prev_stride = out_strides[layers[0]] 165 | elif len(layers) == 2: 166 | assert(layers[0] == ind - 1) 167 | prev_filters = out_filters[layers[0]] + out_filters[layers[1]] 168 | prev_stride = out_strides[layers[0]] 169 | out_filters.append(prev_filters) 170 | out_strides.append(prev_stride) 171 | models.append(EmptyModule()) 172 | elif block['type'] == 'shortcut': 173 | ind = len(models) 174 | prev_filters = out_filters[ind-1] 175 | out_filters.append(prev_filters) 176 | prev_stride = out_strides[ind-1] 177 | out_strides.append(prev_stride) 178 | models.append(EmptyModule()) 179 | elif block['type'] == 'yolo': 180 | yolo_layer = YoloLayer() 181 | anchors = block['anchors'].split(',') 182 | anchor_mask = block['mask'].split(',') 183 | yolo_layer.anchor_mask = [int(i) for i in anchor_mask] 184 | yolo_layer.anchors = [float(i) for i in anchors] 185 | yolo_layer.num_classes = int(block['classes']) 186 | yolo_layer.num_anchors = int(block['num']) 187 | yolo_layer.anchor_step = len(yolo_layer.anchors)//yolo_layer.num_anchors 188 | yolo_layer.stride = prev_stride 189 | out_filters.append(prev_filters) 190 | out_strides.append(prev_stride) 191 | models.append(yolo_layer) 192 | else: 193 | print('unknown type %s' % (block['type'])) 194 | 195 | return models 196 | 197 | def load_weights(self, weightfile): 198 | print() 199 | fp = open(weightfile, 'rb') 200 | header = np.fromfile(fp, count=5, dtype=np.int32) 201 | self.header = torch.from_numpy(header) 202 | self.seen = self.header[3] 203 | buf = np.fromfile(fp, dtype = np.float32) 204 | fp.close() 205 | 206 | start = 0 207 | ind = -2 208 | counter = 3 209 | for block in self.blocks: 210 | if start >= buf.size: 211 | break 212 | ind = ind + 1 213 | if block['type'] == 'net': 214 | continue 215 | elif block['type'] == 'convolutional': 216 | model = self.models[ind] 217 | batch_normalize = int(block['batch_normalize']) 218 | if batch_normalize: 219 | start = load_conv_bn(buf, start, model[0], model[1]) 220 | else: 221 | start = load_conv(buf, start, model[0]) 222 | elif block['type'] == 'upsample': 223 | pass 224 | elif block['type'] == 'route': 225 | pass 226 | elif block['type'] == 'shortcut': 227 | pass 228 | elif block['type'] == 'yolo': 229 | pass 230 | else: 231 | print('unknown type %s' % (block['type'])) 232 | 233 | percent_comp = (counter / len(self.blocks)) * 100 234 | 235 | print('Loading weights. Please Wait...{:.2f}% Complete'.format(percent_comp), end = '\r', flush = True) 236 | 237 | counter += 1 238 | 239 | 240 | 241 | def convert2cpu(gpu_matrix): 242 | return torch.FloatTensor(gpu_matrix.size()).copy_(gpu_matrix) 243 | 244 | 245 | def convert2cpu_long(gpu_matrix): 246 | return torch.LongTensor(gpu_matrix.size()).copy_(gpu_matrix) 247 | 248 | 249 | def get_region_boxes(output, conf_thresh, num_classes, anchors, num_anchors, only_objectness = 1, validation = False): 250 | anchor_step = len(anchors)//num_anchors 251 | if output.dim() == 3: 252 | output = output.unsqueeze(0) 253 | batch = output.size(0) 254 | assert(output.size(1) == (5+num_classes)*num_anchors) 255 | h = output.size(2) 256 | w = output.size(3) 257 | 258 | all_boxes = [] 259 | output = output.view(batch*num_anchors, 5+num_classes, h*w).transpose(0,1).contiguous().view(5+num_classes, batch*num_anchors*h*w) 260 | 261 | grid_x = torch.linspace(0, w-1, w).repeat(h,1).repeat(batch*num_anchors, 1, 1).view(batch*num_anchors*h*w).type_as(output) #cuda() 262 | grid_y = torch.linspace(0, h-1, h).repeat(w,1).t().repeat(batch*num_anchors, 1, 1).view(batch*num_anchors*h*w).type_as(output) #cuda() 263 | xs = torch.sigmoid(output[0]) + grid_x 264 | ys = torch.sigmoid(output[1]) + grid_y 265 | 266 | anchor_w = torch.Tensor(anchors).view(num_anchors, anchor_step).index_select(1, torch.LongTensor([0])) 267 | anchor_h = torch.Tensor(anchors).view(num_anchors, anchor_step).index_select(1, torch.LongTensor([1])) 268 | anchor_w = anchor_w.repeat(batch, 1).repeat(1, 1, h*w).view(batch*num_anchors*h*w).type_as(output) #cuda() 269 | anchor_h = anchor_h.repeat(batch, 1).repeat(1, 1, h*w).view(batch*num_anchors*h*w).type_as(output) #cuda() 270 | ws = torch.exp(output[2]) * anchor_w 271 | hs = torch.exp(output[3]) * anchor_h 272 | 273 | det_confs = torch.sigmoid(output[4]) 274 | cls_confs = torch.nn.Softmax(dim=1)(output[5:5+num_classes].transpose(0,1)).detach() 275 | cls_max_confs, cls_max_ids = torch.max(cls_confs, 1) 276 | cls_max_confs = cls_max_confs.view(-1) 277 | cls_max_ids = cls_max_ids.view(-1) 278 | 279 | 280 | sz_hw = h*w 281 | sz_hwa = sz_hw*num_anchors 282 | det_confs = convert2cpu(det_confs) 283 | cls_max_confs = convert2cpu(cls_max_confs) 284 | cls_max_ids = convert2cpu_long(cls_max_ids) 285 | xs = convert2cpu(xs) 286 | ys = convert2cpu(ys) 287 | ws = convert2cpu(ws) 288 | hs = convert2cpu(hs) 289 | if validation: 290 | cls_confs = convert2cpu(cls_confs.view(-1, num_classes)) 291 | 292 | for b in range(batch): 293 | boxes = [] 294 | for cy in range(h): 295 | for cx in range(w): 296 | for i in range(num_anchors): 297 | ind = b*sz_hwa + i*sz_hw + cy*w + cx 298 | det_conf = det_confs[ind] 299 | if only_objectness: 300 | conf = det_confs[ind] 301 | else: 302 | conf = det_confs[ind] * cls_max_confs[ind] 303 | 304 | if conf > conf_thresh: 305 | bcx = xs[ind] 306 | bcy = ys[ind] 307 | bw = ws[ind] 308 | bh = hs[ind] 309 | cls_max_conf = cls_max_confs[ind] 310 | cls_max_id = cls_max_ids[ind] 311 | box = [bcx/w, bcy/h, bw/w, bh/h, det_conf, cls_max_conf, cls_max_id] 312 | if (not only_objectness) and validation: 313 | for c in range(num_classes): 314 | tmp_conf = cls_confs[ind][c] 315 | if c != cls_max_id and det_confs[ind]*tmp_conf > conf_thresh: 316 | box.append(tmp_conf) 317 | box.append(c) 318 | boxes.append(box) 319 | all_boxes.append(boxes) 320 | 321 | return all_boxes 322 | 323 | 324 | def parse_cfg(cfgfile): 325 | blocks = [] 326 | fp = open(cfgfile, 'r') 327 | block = None 328 | line = fp.readline() 329 | while line != '': 330 | line = line.rstrip() 331 | if line == '' or line[0] == '#': 332 | line = fp.readline() 333 | continue 334 | elif line[0] == '[': 335 | if block: 336 | blocks.append(block) 337 | block = dict() 338 | block['type'] = line.lstrip('[').rstrip(']') 339 | # set default value 340 | if block['type'] == 'convolutional': 341 | block['batch_normalize'] = 0 342 | else: 343 | key,value = line.split('=') 344 | key = key.strip() 345 | if key == 'type': 346 | key = '_type' 347 | value = value.strip() 348 | block[key] = value 349 | line = fp.readline() 350 | 351 | if block: 352 | blocks.append(block) 353 | fp.close() 354 | return blocks 355 | 356 | 357 | def print_cfg(blocks): 358 | print('layer filters size input output'); 359 | prev_width = 416 360 | prev_height = 416 361 | prev_filters = 3 362 | out_filters =[] 363 | out_widths =[] 364 | out_heights =[] 365 | ind = -2 366 | for block in blocks: 367 | ind = ind + 1 368 | if block['type'] == 'net': 369 | prev_width = int(block['width']) 370 | prev_height = int(block['height']) 371 | continue 372 | elif block['type'] == 'convolutional': 373 | filters = int(block['filters']) 374 | kernel_size = int(block['size']) 375 | stride = int(block['stride']) 376 | is_pad = int(block['pad']) 377 | pad = (kernel_size-1)//2 if is_pad else 0 378 | width = (prev_width + 2*pad - kernel_size)//stride + 1 379 | height = (prev_height + 2*pad - kernel_size)//stride + 1 380 | print('%5d %-6s %4d %d x %d / %d %3d x %3d x%4d -> %3d x %3d x%4d' % (ind, 'conv', filters, kernel_size, kernel_size, stride, prev_width, prev_height, prev_filters, width, height, filters)) 381 | prev_width = width 382 | prev_height = height 383 | prev_filters = filters 384 | out_widths.append(prev_width) 385 | out_heights.append(prev_height) 386 | out_filters.append(prev_filters) 387 | elif block['type'] == 'upsample': 388 | stride = int(block['stride']) 389 | filters = prev_filters 390 | width = prev_width*stride 391 | height = prev_height*stride 392 | print('%5d %-6s * %d %3d x %3d x%4d -> %3d x %3d x%4d' % (ind, 'upsample', stride, prev_width, prev_height, prev_filters, width, height, filters)) 393 | prev_width = width 394 | prev_height = height 395 | prev_filters = filters 396 | out_widths.append(prev_width) 397 | out_heights.append(prev_height) 398 | out_filters.append(prev_filters) 399 | elif block['type'] == 'route': 400 | layers = block['layers'].split(',') 401 | layers = [int(i) if int(i) > 0 else int(i)+ind for i in layers] 402 | if len(layers) == 1: 403 | print('%5d %-6s %d' % (ind, 'route', layers[0])) 404 | prev_width = out_widths[layers[0]] 405 | prev_height = out_heights[layers[0]] 406 | prev_filters = out_filters[layers[0]] 407 | elif len(layers) == 2: 408 | print('%5d %-6s %d %d' % (ind, 'route', layers[0], layers[1])) 409 | prev_width = out_widths[layers[0]] 410 | prev_height = out_heights[layers[0]] 411 | assert(prev_width == out_widths[layers[1]]) 412 | assert(prev_height == out_heights[layers[1]]) 413 | prev_filters = out_filters[layers[0]] + out_filters[layers[1]] 414 | out_widths.append(prev_width) 415 | out_heights.append(prev_height) 416 | out_filters.append(prev_filters) 417 | elif block['type'] in ['region', 'yolo']: 418 | print('%5d %-6s' % (ind, 'detection')) 419 | out_widths.append(prev_width) 420 | out_heights.append(prev_height) 421 | out_filters.append(prev_filters) 422 | elif block['type'] == 'shortcut': 423 | from_id = int(block['from']) 424 | from_id = from_id if from_id > 0 else from_id+ind 425 | print('%5d %-6s %d' % (ind, 'shortcut', from_id)) 426 | prev_width = out_widths[from_id] 427 | prev_height = out_heights[from_id] 428 | prev_filters = out_filters[from_id] 429 | out_widths.append(prev_width) 430 | out_heights.append(prev_height) 431 | out_filters.append(prev_filters) 432 | else: 433 | print('unknown type %s' % (block['type'])) 434 | 435 | 436 | def load_conv(buf, start, conv_model): 437 | num_w = conv_model.weight.numel() 438 | num_b = conv_model.bias.numel() 439 | conv_model.bias.data.copy_(torch.from_numpy(buf[start:start+num_b])); start = start + num_b 440 | conv_model.weight.data.copy_(torch.from_numpy(buf[start:start+num_w]).view_as(conv_model.weight.data)); start = start + num_w 441 | return start 442 | 443 | 444 | def load_conv_bn(buf, start, conv_model, bn_model): 445 | num_w = conv_model.weight.numel() 446 | num_b = bn_model.bias.numel() 447 | bn_model.bias.data.copy_(torch.from_numpy(buf[start:start+num_b])); start = start + num_b 448 | bn_model.weight.data.copy_(torch.from_numpy(buf[start:start+num_b])); start = start + num_b 449 | bn_model.running_mean.copy_(torch.from_numpy(buf[start:start+num_b])); start = start + num_b 450 | bn_model.running_var.copy_(torch.from_numpy(buf[start:start+num_b])); start = start + num_b 451 | conv_model.weight.data.copy_(torch.from_numpy(buf[start:start+num_w]).view_as(conv_model.weight.data)); start = start + num_w 452 | return start 453 | -------------------------------------------------------------------------------- /detect_image/migrations/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/atharva-18/Object-Detection-API/6286a3f0477b18da4ed0704be9b224b8210470e7/detect_image/migrations/__init__.py -------------------------------------------------------------------------------- /detect_image/static/css/main.css: -------------------------------------------------------------------------------- 1 | body { 2 | background-color: #30384C; 3 | } 4 | 5 | .main-container { 6 | background-color: #F5F5F1; 7 | height: 550px; 8 | box-shadow: 0px 0px 50px 0px rgba(0, 0, 0, 0.95); 9 | margin-top: 50px !important; 10 | margin: 50px; 11 | position: relative; 12 | } 13 | 14 | .dd-container, 15 | .uf-container, 16 | .dt-container { 17 | position: absolute; 18 | top: 0; 19 | left: 0; 20 | height: inherit; 21 | } 22 | 23 | .dt-container { 24 | z-index: 1; 25 | opacity: 0; 26 | } 27 | 28 | .uf-container { 29 | z-index: 1; 30 | opacity: 0; 31 | } 32 | 33 | .dd-container { 34 | z-index: 2; 35 | } 36 | 37 | .dd-container { 38 | position: center !important; 39 | left: 0; 40 | height: inherit; 41 | } 42 | 43 | .dd-card, 44 | .uf-card, 45 | .dt-card { 46 | height: inherit; 47 | top: -8px !important; 48 | } 49 | 50 | .dd-card { 51 | background-color: #F5F5F1 !important; 52 | border-radius: 8px; 53 | } 54 | 55 | .dd-card { 56 | background-color: #F5F5F1 !important; 57 | border-radius: 8px; 58 | top: -8px !important; 59 | } 60 | 61 | .dt-card { 62 | overflow-y: auto; 63 | } 64 | 65 | #drop-container { 66 | height: 500px; 67 | border-style: solid; 68 | border-color: #34495E; 69 | border-radius: 4px; 70 | } 71 | 72 | .upload-icon { 73 | color: #34495E; 74 | position: absolute; 75 | top: 50%; 76 | left: 50%; 77 | transform: translate(-50%, -125%); 78 | } 79 | 80 | .drag-txt { 81 | position: absolute; 82 | top: 50%; 83 | left: 50%; 84 | transform: translate(-50%, -50%); 85 | font-size: 1.5em; 86 | } 87 | 88 | .or-txt { 89 | position: absolute; 90 | top: 50%; 91 | left: 50%; 92 | transform: translate(-50%, 75%); 93 | } 94 | 95 | #browse-container { 96 | position: absolute; 97 | top: 50%; 98 | left: 50%; 99 | transform: translate(-50%, 100%); 100 | } 101 | 102 | #upload-button { 103 | background-color: #1ABC9C; 104 | position: absolute; 105 | width: calc(50% - 20px); 106 | } 107 | 108 | #go-back { 109 | background-color: #1ABC9C; 110 | position: absolute; 111 | width: calc(50% - 20px); 112 | transform: translate(calc(100% + 20px), 0%); 113 | } 114 | 115 | #go-start { 116 | background-color: #1ABC9C; 117 | position: absolute; 118 | width: calc(100% - 20px); 119 | } 120 | 121 | #img-card-1, 122 | #img-card-2 { 123 | max-width: 100%; 124 | max-height: 100%; 125 | display: block; 126 | } 127 | 128 | .crop { 129 | object-fit: cover; 130 | object-position: center; 131 | width: 100%; 132 | height: 100%; 133 | } 134 | 135 | .loading-text { 136 | margin-left: 120px; 137 | transform: translateY(50%); 138 | } 139 | 140 | .bar { 141 | height: 20px; 142 | background-color: #34495E !important; 143 | } 144 | 145 | .progress { 146 | height: 20px !important; 147 | margin-top: 40px !important; 148 | background-color: #F5F5F5 !important; 149 | } 150 | 151 | .progress .determinate { 152 | background-color: #34495E !important; 153 | } 154 | 155 | .stat-card { 156 | padding-top: 10px !important; 157 | padding-bottom: 10px !important; 158 | } 159 | 160 | .capitalize { 161 | text-transform: capitalize; 162 | } 163 | 164 | .uppercase { 165 | text-transform: uppercase; 166 | } 167 | 168 | -------------------------------------------------------------------------------- /detect_image/static/js/main.js: -------------------------------------------------------------------------------- 1 | $(document).ready(function() { 2 | 3 | var dropContainer = document.getElementById('drop-container'); 4 | dropContainer.ondragover = dropContainer.ondragend = function() { 5 | return false; 6 | }; 7 | 8 | dropContainer.ondrop = function(e) { 9 | e.preventDefault(); 10 | loadImage(e.dataTransfer.files[0]) 11 | } 12 | 13 | $("#browse-button").change(function() { 14 | loadImage($("#browse-button").prop("files")[0]); 15 | }); 16 | 17 | $('.modal').modal({ 18 | dismissible: false, 19 | ready: function(modal, trigger) { 20 | $.ajax({ 21 | type: "POST", 22 | url: '/detect/api/', 23 | data: { 24 | 'image64': $('#img-card-1').attr('src') 25 | }, 26 | dataType: 'text', 27 | success: function(data) { 28 | loadStats(data) 29 | } 30 | }).always(function() { 31 | modal.modal('close'); 32 | }); 33 | } 34 | }); 35 | 36 | $('#go-back').click(function() { 37 | $('#img-card-1').removeAttr("src"); 38 | $('#stat-table').html(''); 39 | switchCard(0); 40 | }); 41 | $('#go-start').click(function() { 42 | var elem = document.getElementById("img-card-2"); 43 | elem.parentNode.removeChild(elem); 44 | $('#stat-table').html(''); 45 | switchCard(0); 46 | }); 47 | 48 | $('#upload-button').click(function() { 49 | $('.modal').modal('open'); 50 | }); 51 | }); 52 | 53 | switchCard = function(cardNo) { 54 | var containers = [".dd-container", ".uf-container", ".dt-container"]; 55 | var visibleContainer = containers[cardNo]; 56 | for (var i = 0; i < containers.length; i++) { 57 | var oz = (containers[i] === visibleContainer) ? '1' : '0'; 58 | $(containers[i]).animate({ 59 | opacity: oz 60 | }, { 61 | duration: 200, 62 | queue: false, 63 | }).css("z-index", oz); 64 | } 65 | } 66 | 67 | loadImage = function(file) { 68 | var reader = new FileReader(); 69 | reader.onload = function(event) { 70 | $('#img-card-1').attr('src', event.target.result); 71 | } 72 | reader.readAsDataURL(file); 73 | switchCard(1); 74 | } 75 | 76 | loadStats = function(jsonData) { 77 | switchCard(2); 78 | var data = JSON.parse(jsonData); 79 | if(data["success"] == true){ 80 | var elem = document.createElement("img"); 81 | elem.setAttribute('class', "card crop"); 82 | elem.setAttribute('id', 'img-card-2'); 83 | elem.src = data['url']; 84 | document.getElementById("result-image").appendChild(elem); 85 | } 86 | } 87 | -------------------------------------------------------------------------------- /detect_image/templates/index.html: -------------------------------------------------------------------------------- 1 | {% load staticfiles %} 2 | 3 | 4 | 5 | 6 | 7 | Object Detection with YOLO 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 |
19 |
20 |
21 |
22 |
23 | cloud_upload 24 | Drag and drop image here 25 | or 26 |
27 |
28 | Browse files 29 | 30 |
31 |
32 |
33 |
34 |
35 |
36 |
37 |
38 | 39 |
40 |
41 | Upload and Detect Image 42 |
43 |
44 | Go Back 45 |
46 |
47 |
48 |
49 |
50 |
51 | Go Back 52 |
53 |
54 | 78 |
79 | 80 | 81 | -------------------------------------------------------------------------------- /detect_image/urls.py: -------------------------------------------------------------------------------- 1 | from django.conf import settings 2 | from django.views.static import serve 3 | 4 | urlpatterns = [ 5 | url(r'^detect_image/(?P.*)$', serve, { 6 | 'document_root': settings.MEDIA_ROOT, 7 | }), 8 | ] -------------------------------------------------------------------------------- /detect_image/utils/__init__.py: -------------------------------------------------------------------------------- 1 | from .utils import * -------------------------------------------------------------------------------- /detect_image/utils/utils.py: -------------------------------------------------------------------------------- 1 | import time 2 | import torch 3 | import numpy as np 4 | import matplotlib.pyplot as plt 5 | import matplotlib.patches as patches 6 | 7 | from django.core.files.temp import NamedTemporaryFile 8 | 9 | import cloudinary 10 | import cloudinary.uploader 11 | import cloudinary.api 12 | 13 | def boxes_iou(box1, box2): 14 | 15 | # Get the Width and Height of each bounding box 16 | width_box1 = box1[2] 17 | height_box1 = box1[3] 18 | width_box2 = box2[2] 19 | height_box2 = box2[3] 20 | 21 | # Calculate the area of the each bounding box 22 | area_box1 = width_box1 * height_box1 23 | area_box2 = width_box2 * height_box2 24 | 25 | # Find the vertical edges of the union of the two bounding boxes 26 | mx = min(box1[0] - width_box1/2.0, box2[0] - width_box2/2.0) 27 | Mx = max(box1[0] + width_box1/2.0, box2[0] + width_box2/2.0) 28 | 29 | # Calculate the width of the union of the two bounding boxes 30 | union_width = Mx - mx 31 | 32 | # Find the horizontal edges of the union of the two bounding boxes 33 | my = min(box1[1] - height_box1/2.0, box2[1] - height_box2/2.0) 34 | My = max(box1[1] + height_box1/2.0, box2[1] + height_box2/2.0) 35 | 36 | # Calculate the height of the union of the two bounding boxes 37 | union_height = My - my 38 | 39 | # Calculate the width and height of the area of intersection of the two bounding boxes 40 | intersection_width = width_box1 + width_box2 - union_width 41 | intersection_height = height_box1 + height_box2 - union_height 42 | 43 | # If the the boxes don't overlap then their IOU is zero 44 | if intersection_width <= 0 or intersection_height <= 0: 45 | return 0.0 46 | 47 | # Calculate the area of intersection of the two bounding boxes 48 | intersection_area = intersection_width * intersection_height 49 | 50 | # Calculate the area of the union of the two bounding boxes 51 | union_area = area_box1 + area_box2 - intersection_area 52 | 53 | # Calculate the IOU 54 | iou = intersection_area/union_area 55 | 56 | return iou 57 | 58 | 59 | def nms(boxes, iou_thresh): 60 | 61 | # If there are no bounding boxes do nothing 62 | if len(boxes) == 0: 63 | return boxes 64 | 65 | # Create a PyTorch Tensor to keep track of the detection confidence 66 | # of each predicted bounding box 67 | det_confs = torch.zeros(len(boxes)) 68 | 69 | # Get the detection confidence of each predicted bounding box 70 | for i in range(len(boxes)): 71 | det_confs[i] = boxes[i][4] 72 | 73 | # Sort the indices of the bounding boxes by detection confidence value in descending order. 74 | # We ignore the first returned element since we are only interested in the sorted indices 75 | _,sortIds = torch.sort(det_confs, descending = True) 76 | 77 | # Create an empty list to hold the best bounding boxes after 78 | # Non-Maximal Suppression (NMS) is performed 79 | best_boxes = [] 80 | 81 | # Perform Non-Maximal Suppression 82 | for i in range(len(boxes)): 83 | 84 | # Get the bounding box with the highest detection confidence first 85 | box_i = boxes[sortIds[i]] 86 | 87 | # Check that the detection confidence is not zero 88 | if box_i[4] > 0: 89 | 90 | # Save the bounding box 91 | best_boxes.append(box_i) 92 | 93 | # Go through the rest of the bounding boxes in the list and calculate their IOU with 94 | # respect to the previous selected box_i. 95 | for j in range(i + 1, len(boxes)): 96 | box_j = boxes[sortIds[j]] 97 | 98 | # If the IOU of box_i and box_j is higher than the given IOU threshold set 99 | # box_j's detection confidence to zero. 100 | if boxes_iou(box_i, box_j) > iou_thresh: 101 | box_j[4] = 0 102 | 103 | return best_boxes 104 | 105 | 106 | def detect_objects(model, img, iou_thresh, nms_thresh): 107 | 108 | # Start the time. This is done to calculate how long the detection takes. 109 | start = time.time() 110 | 111 | # Set the model to evaluation mode. 112 | model.eval() 113 | 114 | # Convert the image from a NumPy ndarray to a PyTorch Tensor of the correct shape. 115 | # The image is transposed, then converted to a FloatTensor of dtype float32, then 116 | # Normalized to values between 0 and 1, and finally unsqueezed to have the correct 117 | # shape of 1 x 3 x 416 x 416 118 | img = torch.from_numpy(img.transpose(2,0,1)).float().div(255.0).unsqueeze(0) 119 | 120 | # Feed the image to the neural network with the corresponding NMS threshold. 121 | # The first step in NMS is to remove all bounding boxes that have a very low 122 | # probability of detection. All predicted bounding boxes with a value less than 123 | # the given NMS threshold will be removed. 124 | list_boxes = model(img, nms_thresh) 125 | 126 | # Make a new list with all the bounding boxes returned by the neural network 127 | boxes = list_boxes[0][0] + list_boxes[1][0] + list_boxes[2][0] 128 | 129 | # Perform the second step of NMS on the bounding boxes returned by the neural network. 130 | # In this step, we only keep the best bounding boxes by eliminating all the bounding boxes 131 | # whose IOU value is higher than the given IOU threshold 132 | boxes = nms(boxes, iou_thresh) 133 | 134 | # Stop the time. 135 | finish = time.time() 136 | 137 | # Print the time it took to detect objects 138 | print('\n\nIt took {:.3f}'.format(finish - start), 'seconds to detect the objects in the image.\n') 139 | 140 | # Print the number of objects detected 141 | print('Number of Objects Detected:', len(boxes), '\n') 142 | 143 | return boxes 144 | 145 | 146 | def load_class_names(namesfile): 147 | 148 | # Create an empty list to hold the object classes 149 | class_names = [] 150 | 151 | # Open the file containing the COCO object classes in read-only mode 152 | with open(namesfile, 'r') as fp: 153 | 154 | # The coco.names file contains only one object class per line. 155 | # Read the file line by line and save all the lines in a list. 156 | lines = fp.readlines() 157 | 158 | # Get the object class names 159 | for line in lines: 160 | 161 | # Make a copy of each line with any trailing whitespace removed 162 | line = line.rstrip() 163 | 164 | # Save the object class name into class_names 165 | class_names.append(line) 166 | 167 | return class_names 168 | 169 | 170 | def print_objects(boxes, class_names): 171 | print('Objects Found and Confidence Level:\n') 172 | result = {} 173 | for i in range(len(boxes)): 174 | box = boxes[i] 175 | if len(box) >= 7 and class_names: 176 | cls_conf = box[5] 177 | cls_id = box[6] 178 | result[class_names[cls_id]] = cls_conf.item() 179 | return result 180 | 181 | 182 | def plot_boxes(img, boxes, class_names, plot_labels, color = None): 183 | 184 | # Define a tensor used to set the colors of the bounding boxes 185 | colors = torch.FloatTensor([[1,0,1],[0,0,1],[0,1,1],[0,1,0],[1,1,0],[1,0,0]]) 186 | 187 | # Define a function to set the colors of the bounding boxes 188 | def get_color(c, x, max_val): 189 | ratio = float(x) / max_val * 5 190 | i = int(np.floor(ratio)) 191 | j = int(np.ceil(ratio)) 192 | 193 | ratio = ratio - i 194 | r = (1 - ratio) * colors[i][c] + ratio * colors[j][c] 195 | 196 | return int(r * 255) 197 | 198 | # Get the width and height of the image 199 | width = img.shape[1] 200 | height = img.shape[0] 201 | 202 | # Create a figure and plot the image 203 | fig, a = plt.subplots(1,1) 204 | a.imshow(img) 205 | 206 | # Plot the bounding boxes and corresponding labels on top of the image 207 | for i in range(len(boxes)): 208 | 209 | # Get the ith bounding box 210 | box = boxes[i] 211 | 212 | # Get the (x,y) pixel coordinates of the lower-left and lower-right corners 213 | # of the bounding box relative to the size of the image. 214 | x1 = int(np.around((box[0] - box[2]/2.0) * width)) 215 | y1 = int(np.around((box[1] - box[3]/2.0) * height)) 216 | x2 = int(np.around((box[0] + box[2]/2.0) * width)) 217 | y2 = int(np.around((box[1] + box[3]/2.0) * height)) 218 | 219 | # Set the default rgb value to red 220 | rgb = (1, 0, 0) 221 | 222 | # Use the same color to plot the bounding boxes of the same object class 223 | if len(box) >= 7 and class_names: 224 | cls_conf = box[5] 225 | cls_id = box[6] 226 | classes = len(class_names) 227 | offset = cls_id * 123457 % classes 228 | red = get_color(2, offset, classes) / 255 229 | green = get_color(1, offset, classes) / 255 230 | blue = get_color(0, offset, classes) / 255 231 | 232 | # If a color is given then set rgb to the given color instead 233 | if color is None: 234 | rgb = (red, green, blue) 235 | else: 236 | rgb = color 237 | 238 | # Calculate the width and height of the bounding box relative to the size of the image. 239 | width_x = x2 - x1 240 | width_y = y1 - y2 241 | 242 | # Set the postion and size of the bounding box. (x1, y2) is the pixel coordinate of the 243 | # lower-left corner of the bounding box relative to the size of the image. 244 | rect = patches.Rectangle((x1, y2), 245 | width_x, width_y, 246 | linewidth = 2, 247 | edgecolor = rgb, 248 | facecolor = 'none') 249 | 250 | # Draw the bounding box on top of the image 251 | a.add_patch(rect) 252 | 253 | # If plot_labels = True then plot the corresponding label 254 | if plot_labels: 255 | 256 | # Create a string with the object class name and the corresponding object class probability 257 | conf_tx = class_names[cls_id] + ': {:.1f}'.format(cls_conf) 258 | 259 | # Define x and y offsets for the labels 260 | lxc = (img.shape[1] * 0.266) / 100 261 | lyc = (img.shape[0] * 1.180) / 100 262 | 263 | # Draw the labels on top of the image 264 | a.text(x1 + lxc, y1 - lyc, conf_tx, fontsize = 12, color = 'k', 265 | bbox = dict(facecolor = rgb, edgecolor = rgb, alpha = 0.8)) 266 | 267 | tmp_plot = NamedTemporaryFile() 268 | plt.axis('off') 269 | plt.savefig(tmp_plot.name, format='png', dpi=360, bbox_inches='tight') 270 | 271 | try: 272 | response = cloudinary.uploader.upload(tmp_plot.name, 273 | public_id = 'result', 274 | overwrite= True) 275 | except: 276 | response = {'url':''} 277 | tmp_plot.close() 278 | 279 | return response['url'] 280 | -------------------------------------------------------------------------------- /detect_image/views.py: -------------------------------------------------------------------------------- 1 | from django.shortcuts import render 2 | from django.http import HttpResponse, JsonResponse 3 | from django.template import loader 4 | from django.views.decorators.csrf import csrf_exempt 5 | from django.core.files.temp import NamedTemporaryFile 6 | 7 | import io 8 | import os 9 | from PIL import Image 10 | import cv2 11 | import numpy as np 12 | from base64 import b64decode, b64encode 13 | from .utils import * 14 | from .darknet import Darknet 15 | 16 | # Create your views here. 17 | ######################### 18 | 19 | @csrf_exempt 20 | def yolo_detect_api(request): 21 | data = {'success':False} 22 | url = '' 23 | 24 | if request.method == "POST": 25 | if request.FILES.get("image", None) is not None: 26 | image_request = request.FILES["image"] 27 | image_bytes = image_request.read() 28 | image = Image.open(io.BytesIO(image_bytes)) 29 | result, url = yolo_detect(image) 30 | elif request.POST.get("image64", None) is not None: 31 | base64_data = request.POST.get("image64", None).split(',', 1)[1] 32 | plain_data = b64decode(base64_data) 33 | plain_data = np.array(Image.open(io.BytesIO(plain_data))) 34 | result, url = yolo_detect(plain_data) 35 | 36 | if result: 37 | data['success'] = True 38 | 39 | data['objects'] = result 40 | data['url'] = url 41 | return JsonResponse(data) 42 | 43 | def detect(request): 44 | return render(request, 'index.html') 45 | 46 | def yolo_detect(original_image): 47 | cfg_file = './cfg/yolov3.cfg' 48 | weight_file = './weights/yolov3.weights' 49 | namesfile = 'data/coco.names' 50 | 51 | m = Darknet(cfg_file) 52 | m.load_weights(weight_file) 53 | class_names = load_class_names(namesfile) 54 | 55 | resized_image = cv2.resize(original_image, (m.width, m.height)) 56 | 57 | nms_thresh = 0.6 58 | iou_thresh = 0.4 59 | 60 | boxes = detect_objects(m, resized_image, iou_thresh, nms_thresh) 61 | url = plot_boxes(original_image, boxes, class_names, plot_labels = True) 62 | objects = print_objects(boxes, class_names) 63 | return objects, url 64 | -------------------------------------------------------------------------------- /manage.py: -------------------------------------------------------------------------------- 1 | #!/usr/bin/env python 2 | """Django's command-line utility for administrative tasks.""" 3 | import os 4 | import sys 5 | 6 | 7 | def main(): 8 | os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'website.settings') 9 | try: 10 | from django.core.management import execute_from_command_line 11 | except ImportError as exc: 12 | raise ImportError( 13 | "Couldn't import Django. Are you sure it's installed and " 14 | "available on your PYTHONPATH environment variable? Did you " 15 | "forget to activate a virtual environment?" 16 | ) from exc 17 | execute_from_command_line(sys.argv) 18 | 19 | 20 | if __name__ == '__main__': 21 | main() 22 | -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- 1 | Django~=2.2.4 2 | Pillow~=6.2.0 3 | gunicorn~=19.9.0 4 | WhiteNoise~=4.1.3 5 | Keras~=2.2.5 6 | numpy~=1.16.4 7 | matplotlib~=3.1.0 8 | https://download.pytorch.org/whl/cpu/torch-0.4.1.post2-cp37-cp37m-linux_x86_64.whl 9 | opencv-python~=4.1.0.25 10 | imageio~=2.5.0 11 | gunicorn~=19.9.0 12 | cloudinary 13 | -------------------------------------------------------------------------------- /result.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/atharva-18/Object-Detection-API/6286a3f0477b18da4ed0704be9b224b8210470e7/result.png -------------------------------------------------------------------------------- /runtime.txt: -------------------------------------------------------------------------------- 1 | python-3.7.3 -------------------------------------------------------------------------------- /temp.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/atharva-18/Object-Detection-API/6286a3f0477b18da4ed0704be9b224b8210470e7/temp.png -------------------------------------------------------------------------------- /website/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/atharva-18/Object-Detection-API/6286a3f0477b18da4ed0704be9b224b8210470e7/website/__init__.py -------------------------------------------------------------------------------- /website/settings.py: -------------------------------------------------------------------------------- 1 | """ 2 | Django settings for website project. 3 | 4 | Generated by 'django-admin startproject' using Django 2.2.4. 5 | 6 | For more information on this file, see 7 | https://docs.djangoproject.com/en/2.2/topics/settings/ 8 | 9 | For the full list of settings and their values, see 10 | https://docs.djangoproject.com/en/2.2/ref/settings/ 11 | """ 12 | 13 | import os 14 | import cloudinary 15 | import cloudinary.uploader 16 | import cloudinary.api 17 | 18 | # Build paths inside the project like this: os.path.join(BASE_DIR, ...) 19 | BASE_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__))) 20 | 21 | 22 | # Quick-start development settings - unsuitable for production 23 | # See https://docs.djangoproject.com/en/2.2/howto/deployment/checklist/ 24 | 25 | # SECURITY WARNING: keep the secret key used in production secret! 26 | SECRET_KEY = 'zvk*2xwi*p1qgxwk_$1ihrbyu&91%7nhryv7!o1nx5_j)2i+0n' 27 | 28 | # SECURITY WARNING: don't run with debug turned on in production! 29 | DEBUG = False 30 | 31 | ALLOWED_HOSTS = [ 32 | '127.0.0.1', 33 | 'object-detection-yolo.herokuapp.com' 34 | ] 35 | 36 | 37 | # Application definition 38 | 39 | INSTALLED_APPS = [ 40 | 'django.contrib.admin', 41 | 'django.contrib.auth', 42 | 'django.contrib.contenttypes', 43 | 'django.contrib.sessions', 44 | 'django.contrib.messages', 45 | 'django.contrib.staticfiles', 46 | 'detect_image.apps.DetectImageConfig', 47 | ] 48 | 49 | MIDDLEWARE = [ 50 | 'django.middleware.security.SecurityMiddleware', 51 | 'whitenoise.middleware.WhiteNoiseMiddleware', 52 | 'django.contrib.sessions.middleware.SessionMiddleware', 53 | 'django.middleware.common.CommonMiddleware', 54 | 'django.middleware.csrf.CsrfViewMiddleware', 55 | 'django.contrib.auth.middleware.AuthenticationMiddleware', 56 | 'django.contrib.messages.middleware.MessageMiddleware', 57 | 'django.middleware.clickjacking.XFrameOptionsMiddleware', 58 | ] 59 | 60 | ROOT_URLCONF = 'website.urls' 61 | 62 | TEMPLATES = [ 63 | { 64 | 'BACKEND': 'django.template.backends.django.DjangoTemplates', 65 | 'DIRS': [], 66 | 'APP_DIRS': True, 67 | 'OPTIONS': { 68 | 'context_processors': [ 69 | 'django.template.context_processors.debug', 70 | 'django.template.context_processors.request', 71 | 'django.contrib.auth.context_processors.auth', 72 | 'django.contrib.messages.context_processors.messages', 73 | ], 74 | }, 75 | }, 76 | ] 77 | 78 | WSGI_APPLICATION = 'website.wsgi.application' 79 | 80 | 81 | # Database 82 | # https://docs.djangoproject.com/en/2.2/ref/settings/#databases 83 | 84 | DATABASES = { 85 | 'default': { 86 | 'ENGINE': 'django.db.backends.sqlite3', 87 | 'NAME': os.path.join(BASE_DIR, 'db.sqlite3'), 88 | } 89 | } 90 | 91 | 92 | # Password validation 93 | # https://docs.djangoproject.com/en/2.2/ref/settings/#auth-password-validators 94 | 95 | AUTH_PASSWORD_VALIDATORS = [ 96 | { 97 | 'NAME': 'django.contrib.auth.password_validation.UserAttributeSimilarityValidator', 98 | }, 99 | { 100 | 'NAME': 'django.contrib.auth.password_validation.MinimumLengthValidator', 101 | }, 102 | { 103 | 'NAME': 'django.contrib.auth.password_validation.CommonPasswordValidator', 104 | }, 105 | { 106 | 'NAME': 'django.contrib.auth.password_validation.NumericPasswordValidator', 107 | }, 108 | ] 109 | 110 | 111 | # Internationalization 112 | # https://docs.djangoproject.com/en/2.2/topics/i18n/ 113 | 114 | LANGUAGE_CODE = 'en-us' 115 | 116 | TIME_ZONE = 'UTC' 117 | 118 | USE_I18N = True 119 | 120 | USE_L10N = True 121 | 122 | USE_TZ = True 123 | 124 | 125 | # Static files (CSS, JavaScript, Images) 126 | # https://docs.djangoproject.com/en/2.2/howto/static-files/ 127 | 128 | STATICFILES_STORAGE = 'whitenoise.storage.CompressedManifestStaticFilesStorage' 129 | 130 | PROJECT_ROOT = os.path.dirname(os.path.abspath(__file__)) 131 | 132 | STATIC_ROOT = os.path.join(BASE_DIR, 'staticfiles') 133 | 134 | STATIC_URL = '/static/' 135 | 136 | # Extra places for collectstatic to find static files. 137 | # STATICFILES_DIRS = [ 138 | # os.path.join(PROJECT_ROOT, 'staticfiles'), 139 | # ] 140 | 141 | MEDIA_ROOT = BASE_DIR 142 | 143 | MEDIA_URL = 'detect/' 144 | 145 | # Put your cloudinary credentials here 146 | 147 | cloudinary.config( 148 | cloud_name = "", 149 | api_key = "", 150 | api_secret = "" 151 | ) 152 | -------------------------------------------------------------------------------- /website/urls.py: -------------------------------------------------------------------------------- 1 | """website URL Configuration 2 | 3 | The `urlpatterns` list routes URLs to views. For more information please see: 4 | https://docs.djangoproject.com/en/2.2/topics/http/urls/ 5 | Examples: 6 | Function views 7 | 1. Add an import: from my_app import views 8 | 2. Add a URL to urlpatterns: path('', views.home, name='home') 9 | Class-based views 10 | 1. Add an import: from other_app.views import Home 11 | 2. Add a URL to urlpatterns: path('', Home.as_view(), name='home') 12 | Including another URLconf 13 | 1. Import the include() function: from django.urls import include, path 14 | 2. Add a URL to urlpatterns: path('blog/', include('blog.urls')) 15 | """ 16 | from django.urls import path 17 | from django.views.generic import RedirectView 18 | from detect_image.views import detect, yolo_detect_api 19 | from django.conf import settings 20 | from django.conf.urls.static import static 21 | 22 | urlpatterns = [ 23 | path('detect/', detect), 24 | path('', RedirectView.as_view(url='detect/')), 25 | path('detect/api/', yolo_detect_api), 26 | ] + static(settings.MEDIA_URL, document_root=settings.MEDIA_ROOT) 27 | -------------------------------------------------------------------------------- /website/wsgi.py: -------------------------------------------------------------------------------- 1 | """ 2 | WSGI config for website project. 3 | 4 | It exposes the WSGI callable as a module-level variable named ``application``. 5 | 6 | For more information on this file, see 7 | https://docs.djangoproject.com/en/2.2/howto/deployment/wsgi/ 8 | """ 9 | 10 | import os 11 | 12 | from django.core.wsgi import get_wsgi_application 13 | 14 | os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'website.settings') 15 | 16 | application = get_wsgi_application() 17 | -------------------------------------------------------------------------------- /weights/w: -------------------------------------------------------------------------------- 1 | 2 | --------------------------------------------------------------------------------