├── .ipynb_checkpoints
    ├── (draft) Face Recognition no bkg-checkpoint.ipynb
    ├── (draft) Face Recognition-checkpoint.ipynb
    ├── Face Recognition with Correlation Coefficients and Euclidean Distance. Comparison of the two methods.-checkpoint.ipynb
    ├── Test-checkpoint.ipynb
    └── Untitled-checkpoint.ipynb
├── Face Recognition with Correlation Coefficients and Euclidean Distance. Comparison of the two methods..ipynb
├── README.md
├── compare.jpg
├── demo_preparation_img.jpg
├── images-faceonly
    ├── js01.jpg
    ├── js02.jpg
    ├── js03.jpg
    ├── js04.jpg
    ├── js05.jpg
    ├── js06.jpg
    ├── js07.jpg
    ├── js08.jpg
    ├── js09.jpg
    └── js10.jpg
├── images-nobkg
    ├── js01.jpg
    ├── js02.jpg
    ├── js03.jpg
    ├── js04.jpg
    ├── js05.jpg
    ├── js06.jpg
    ├── js07.jpg
    ├── js08.jpg
    ├── js09.jpg
    └── js10.jpg
├── images-test-faceonly
    ├── test_01_js.jpg
    ├── test_02_js.jpg
    ├── test_03_js.jpg
    ├── test_04_js.jpg
    ├── test_05_ws.jpg
    ├── test_06_kr.jpg
    ├── test_07_jc.jpg
    └── test_08_ip.jpg
├── images-test-nobkg
    ├── test_01_js.jpg
    ├── test_02_js.jpg
    ├── test_03_js.jpg
    ├── test_04_js.jpg
    ├── test_05_ws.jpg
    ├── test_06_kr.jpg
    ├── test_07_jc.jpg
    └── test_08_ip.jpg
├── images-test
    ├── test_01_js.jpg
    ├── test_02_js.jpg
    ├── test_03_js.jpg
    ├── test_04_js.jpg
    ├── test_05_ws.jpg
    ├── test_06_kr.jpg
    ├── test_07_jc.jpg
    └── test_08_ip.jpg
└── images
    ├── js01.jpg
    ├── js02.jpg
    ├── js03.jpg
    ├── js04.jpg
    ├── js05.jpg
    ├── js06.jpg
    ├── js07.jpg
    ├── js08.jpg
    ├── js09.jpg
    └── js10.jpg


/.ipynb_checkpoints/Test-checkpoint.ipynb:
--------------------------------------------------------------------------------
1 | {
2 |  "cells": [],
3 |  "metadata": {},
4 |  "nbformat": 4,
5 |  "nbformat_minor": 5
6 | }
7 | 


--------------------------------------------------------------------------------
/.ipynb_checkpoints/Untitled-checkpoint.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "code",
  5 |    "execution_count": 14,
  6 |    "id": "22e95cbb",
  7 |    "metadata": {},
  8 |    "outputs": [],
  9 |    "source": [
 10 |     "import numpy as np\n",
 11 |     "from PIL import Image\n",
 12 |     "from sklearn.decomposition import PCA\n",
 13 |     "import os\n",
 14 |     "import matplotlib.pyplot as plt"
 15 |    ]
 16 |   },
 17 |   {
 18 |    "cell_type": "code",
 19 |    "execution_count": 15,
 20 |    "id": "9525f587",
 21 |    "metadata": {},
 22 |    "outputs": [
 23 |     {
 24 |      "data": {
 25 |       "image/png": "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\n",
 26 |       "text/plain": [
 27 |        "<Figure size 640x480 with 1 Axes>"
 28 |       ]
 29 |      },
 30 |      "metadata": {},
 31 |      "output_type": "display_data"
 32 |     }
 33 |    ],
 34 |    "source": [
 35 |     "\n",
 36 |     "\n",
 37 |     "# Set directory path to images folder\n",
 38 |     "directory = \"images-b/\"\n",
 39 |     "\n",
 40 |     "# Create empty array to store images\n",
 41 |     "image_array = np.zeros((5, 200 * 200))\n",
 42 |     "\n",
 43 |     "# Load images into array\n",
 44 |     "for i, filename in enumerate(os.listdir(directory)):\n",
 45 |     "    if filename.endswith(\".jpg\"):\n",
 46 |     "        img = Image.open(directory + filename).convert('L')  # Convert to grayscale\n",
 47 |     "        img_array = np.array(img)\n",
 48 |     "        image_array[i] = img_array.flatten()\n",
 49 |     "\n",
 50 |     "# Normalize data by centering around the mean pixel value\n",
 51 |     "data_centered = image_array - np.mean(image_array, axis=0)\n",
 52 |     "\n",
 53 |     "# Compute eigenfaces using PCA\n",
 54 |     "pca = PCA(n_components=5, svd_solver='full')\n",
 55 |     "pca.fit(data_centered)\n",
 56 |     "eigenfaces = pca.components_\n",
 57 |     "\n",
 58 |     "# Display first eigenface\n",
 59 |     "\n",
 60 |     "plt.imshow(eigenfaces[0].reshape((200, 200)), cmap='gray')\n",
 61 |     "plt.show()"
 62 |    ]
 63 |   },
 64 |   {
 65 |    "cell_type": "markdown",
 66 |    "id": "03035e19",
 67 |    "metadata": {},
 68 |    "source": [
 69 |     "#### да видя как да сравня тази снимка с други мои. с други на други хора, някви известни, да видя какъв процент съвпадение ще намери. да видя дали може да ми открие лицето в голяма снимка\n",
 70 |     "\n",
 71 |     "\n",
 72 |     "#### Да пробвам как работи \"открийте разликите\" за разлики между изображения от нета, като извадя един np.array от друг\n",
 73 |     "\n",
 74 |     "да направя проверка, че матриците ми са с една и съща размерност\n",
 75 |     "да видя какви други проверки мога да направя"
 76 |    ]
 77 |   },
 78 |   {
 79 |    "cell_type": "markdown",
 80 |    "id": "8525febf",
 81 |    "metadata": {},
 82 |    "source": [
 83 |     "Основната математическа формула, която се използва в кода, е методът на главните компоненти (PCA) за извличане на основните признаци от множество данни, които имат множество корелации. Този метод се използва за намаляване на размерността на многомерни данни, като се представят чрез компоненти, които са линейни комбинации на входните данни.\n",
 84 |     "\n",
 85 |     "Формулата за PCA е:\n",
 86 |     "\n",
 87 |     "$X = U \\Sigma V^T$\n",
 88 |     "\n",
 89 |     "където $X$ е матрицата на данните, $U$ е матрицата на главните компоненти, $\\Sigma$ е матрицата на сингулярните стойности, а $V$ е матрицата на главните компоненти на $X^T$.\n",
 90 |     "\n",
 91 |     "Друга формула, която се използва, е формулата за нормализация на данни, която се прилага за центриране на данните около средната им стойност. Тази формула е:\n",
 92 |     "\n",
 93 |     "$\\frac{x - \\mu}{\\sigma}$\n",
 94 |     "\n",
 95 |     "където $x$ е стойността на данните, $\\mu$ е средната стойност на данните, а $\\sigma$ е стандартното отклонение на данните."
 96 |    ]
 97 |   },
 98 |   {
 99 |    "cell_type": "code",
100 |    "execution_count": 10,
101 |    "id": "2399c0ba",
102 |    "metadata": {},
103 |    "outputs": [
104 |     {
105 |      "name": "stdout",
106 |      "output_type": "stream",
107 |      "text": [
108 |       "Euclidean distance between test images: 14653.11\n"
109 |      ]
110 |     }
111 |    ],
112 |    "source": [
113 |     "\n",
114 |     "# Set directory path to images folder\n",
115 |     "directory = \"images-b/\"\n",
116 |     "\n",
117 |     "# Create empty array to store images\n",
118 |     "image_array = np.zeros((5, 200 * 200))\n",
119 |     "\n",
120 |     "# Load images into array\n",
121 |     "for i, filename in enumerate(os.listdir(directory)):\n",
122 |     "    if filename.endswith(\".jpg\"):\n",
123 |     "        img = Image.open(directory + filename).convert('L')  # Convert to grayscale\n",
124 |     "        img_array = np.array(img)\n",
125 |     "        image_array[i] = img_array.flatten()\n",
126 |     "\n",
127 |     "# Normalize data by centering around the mean pixel value\n",
128 |     "data_centered = image_array - np.mean(image_array, axis=0)\n",
129 |     "\n",
130 |     "# Compute eigenfaces using PCA\n",
131 |     "pca = PCA(n_components=5, svd_solver='full')\n",
132 |     "pca.fit(data_centered)\n",
133 |     "eigenfaces = pca.components_\n",
134 |     "\n",
135 |     "# Load test images\n",
136 |     "test_image1 = np.array(Image.open('images-b/ivo_200x200_1.jpg').convert('L')).flatten()\n",
137 |     "test_image2 = np.array(Image.open('images-b/ivo_200x200_2.jpg').convert('L')).flatten()\n",
138 |     "\n",
139 |     "# Normalize test images by centering around the mean pixel value\n",
140 |     "test_image1_normalized = test_image1 - np.mean(image_array, axis=0)\n",
141 |     "test_image2_normalized = test_image2 - np.mean(image_array, axis=0)\n",
142 |     "\n",
143 |     "# Project test images onto eigenfaces\n",
144 |     "test_image1_coefficients = np.dot(test_image1_normalized, eigenfaces.T)\n",
145 |     "test_image2_coefficients = np.dot(test_image2_normalized, eigenfaces.T)\n",
146 |     "\n",
147 |     "# Calculate Euclidean distance between test image coefficients\n",
148 |     "distance = np.linalg.norm(test_image1_coefficients - test_image2_coefficients)\n",
149 |     "\n",
150 |     "print(f\"Euclidean distance between test images: {distance:.2f}\")"
151 |    ]
152 |   },
153 |   {
154 |    "cell_type": "code",
155 |    "execution_count": 13,
156 |    "id": "63afc7cc",
157 |    "metadata": {},
158 |    "outputs": [
159 |     {
160 |      "ename": "NameError",
161 |      "evalue": "name 'data' is not defined",
162 |      "output_type": "error",
163 |      "traceback": [
164 |       "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
165 |       "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
166 |       "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_21924\\3518195775.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      7\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      8\u001b[0m \u001b[1;31m# Center test image\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 9\u001b[1;33m \u001b[0mtest_image_centered\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtest_image_flat\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmean\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     10\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     11\u001b[0m \u001b[1;31m# Project test image onto eigenfaces\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
167 |       "\u001b[1;31mNameError\u001b[0m: name 'data' is not defined"
168 |      ]
169 |     }
170 |    ],
171 |    "source": [
172 |     "# Load image to compare with eigenfaces\n",
173 |     "test_image = plt.imread('images-b/ivo_200x200_1.jpg')\n",
174 |     "test_image = test_image[:,:,0]\n",
175 |     "\n",
176 |     "# Flatten test image into a 1D array\n",
177 |     "test_image_flat = test_image.flatten()\n",
178 |     "\n",
179 |     "# Center test image\n",
180 |     "test_image_centered = test_image_flat - np.mean(data, axis=0)\n",
181 |     "\n",
182 |     "# Project test image onto eigenfaces\n",
183 |     "test_image_weights = np.dot(test_image_centered, eigenfaces.T)\n",
184 |     "\n",
185 |     "# Reconstruct test image from weights and eigenfaces\n",
186 |     "test_image_reconstructed = np.dot(test_image_weights, eigenfaces) + average_face\n",
187 |     "\n",
188 |     "# Compute cosine similarity between test image and average face\n",
189 |     "cosine_sim = np.dot(test_image_centered, average_face) / (np.linalg.norm(test_image_centered) * np.linalg.norm(average_face))\n",
190 |     "\n",
191 |     "# Display test image and similarity score\n",
192 |     "fig, ax = plt.subplots(1, 2)\n",
193 |     "ax[0].imshow(test_image, cmap='gray')\n",
194 |     "ax[0].set_title('Test Image')\n",
195 |     "ax[1].text(0.5, 0.5, f'Cosine Similarity: {cosine_sim:.4f}', ha='center', va='center')\n",
196 |     "ax[1].axis('off')\n",
197 |     "plt.show()"
198 |    ]
199 |   },
200 |   {
201 |    "cell_type": "code",
202 |    "execution_count": null,
203 |    "id": "5f738607",
204 |    "metadata": {},
205 |    "outputs": [],
206 |    "source": []
207 |   },
208 |   {
209 |    "cell_type": "code",
210 |    "execution_count": null,
211 |    "id": "48deded6",
212 |    "metadata": {},
213 |    "outputs": [],
214 |    "source": []
215 |   }
216 |  ],
217 |  "metadata": {
218 |   "kernelspec": {
219 |    "display_name": "Python 3 (ipykernel)",
220 |    "language": "python",
221 |    "name": "python3"
222 |   },
223 |   "language_info": {
224 |    "codemirror_mode": {
225 |     "name": "ipython",
226 |     "version": 3
227 |    },
228 |    "file_extension": ".py",
229 |    "mimetype": "text/x-python",
230 |    "name": "python",
231 |    "nbconvert_exporter": "python",
232 |    "pygments_lexer": "ipython3",
233 |    "version": "3.9.13"
234 |   }
235 |  },
236 |  "nbformat": 4,
237 |  "nbformat_minor": 5
238 | }
239 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
 1 | # Face recognition with correlation coefficients and euclidean distance. comparison of the two methods.
 2 | 
 3 | ## Introduction
 4 | 
 5 | The goal of this project is based on photos of a person that we have and know to be his, to establish new test photos whether they are of the same person or not.
 6 | 
 7 | Facial recognition has a number of applications
 8 | - The realm of security. Identifying a perpetrator of a crime in a photo database.
 9 | - Verification in apps with facial recognition
10 | - Unlock phones or laptops with facial recognition
11 | 
12 | The goal of this project is based on photos of a person that we have and know to be his, to establish new test photos whether they are of the same person or not.
13 | 
14 | We will compare images by Correlation Coefficients as well as by Euclidean Distance. Presenting their results, we will find out which method gives correct results. We will then compare the performance of the two methods by graphing it and then measuring it.
15 | 
16 | 
17 | ## Raw Data
18 | 
19 | For the purposes of the experiment, 10 photos of the actor Jason Statham were collected from the Internet, setting an averaged image.
20 | 
21 | <tr>
22 | <td><img src="images-faceonly/js01.jpg" width="200" height="200" />
23 | <td><img src="images-faceonly/js02.jpg" width="200" height="200" />
24 | <td><img src="images-faceonly/js03.jpg" width="200" height="200" />
25 | <td><img src="images-faceonly/js04.jpg" width="200" height="200" />
26 | <td><img src="images-faceonly/js05.jpg" width="200" height="200" />
27 | </tr>
28 | <tr>
29 | <td><img src="images-faceonly/js06.jpg" width="200" height="200" />
30 | <td><img src="images-faceonly/js07.jpg" width="200" height="200" />
31 | <td><img src="images-faceonly/js08.jpg" width="200" height="200" />
32 | <td><img src="images-faceonly/js09.jpg" width="200" height="200" />
33 | <td><img src="images-faceonly/js10.jpg" width="200" height="200" />
34 | </tr>
35 | 
36 | </table>
37 | 
38 | 
39 | Images of other faces are also collected to compare with the averaged one. 4 of them to the same person and 4 to other persons.
40 | 
41 | <table>
42 | <tr>
43 | <td><img src="images-test-faceonly/test_01_js.jpg" width="200" height="200" />
44 | <td><img src="images-test-faceonly/test_02_js.jpg" width="200" height="200" />
45 | <td><img src="images-test-faceonly/test_03_js.jpg" width="200" height="200" />
46 | <td><img src="images-test-faceonly/test_04_js.jpg" width="200" height="200" />
47 | <td><img src="images-test-faceonly/test_05_ws.jpg" width="200" height="200" />
48 | <td><img src="images-test-faceonly/test_06_kr.jpg" width="200" height="200" />
49 | <td><img src="images-test-faceonly/test_07_jc.jpg" width="200" height="200" />
50 | <td><img src="images-test-faceonly/test_08_ip.jpg" width="200" height="200" />
51 | </tr>
52 | <tr>
53 | <td>test_01_js
54 | <td>test_02_js
55 | <td>test_03_js
56 | <td>test_04_js
57 | <td>test_05_ws
58 | <td>test_06_kr
59 | <td>test_07_jc
60 | <td>test_08_ip
61 | </tr>
62 | <tr>
63 | <td>Jason Statham
64 | <td>Jason Statham
65 | <td>Jason Statham
66 | <td>Jason Statham
67 | <td>Will Smith
68 | <td>Kiano Rives
69 | <td>Jekie Chan
70 | <td>Ivo Petkov (Me)
71 | </tr>
72 | </table>
73 | 
74 | ## Euclidean distance
75 | 
76 | $$\text{distance} = \sqrt{\sum_{i=1}^{n} (x_i - y_i)^2}\$$
77 | 
78 | 
79 | ## Correlation coefficients
80 | 
81 | $$r = \frac{{}\sum_{i=1}^{n} (x_i - \overline{x})(y_i - \overline{y})}
82 | {\sqrt{\sum_{i=1}^{n} (x_i - \overline{x})^2 \sum_{i=1}^{n}(y_i - \overline{y})^2}} $$
83 | 
84 | 
85 | 
86 | ## Calculated comparison of the results of the two methods
87 | To compare which of the two methods more clearly distinguishes Jason Statham's images from the others, we will check the boundaries of the two models' values that lie between the two sets of photos.
88 | 
89 | 
90 | ![Compare](compare.jpg)
91 | 
92 | 
93 | 


--------------------------------------------------------------------------------
/compare.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/compare.jpg


--------------------------------------------------------------------------------
/demo_preparation_img.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/demo_preparation_img.jpg


--------------------------------------------------------------------------------
/images-faceonly/js01.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-faceonly/js01.jpg


--------------------------------------------------------------------------------
/images-faceonly/js02.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-faceonly/js02.jpg


--------------------------------------------------------------------------------
/images-faceonly/js03.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-faceonly/js03.jpg


--------------------------------------------------------------------------------
/images-faceonly/js04.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-faceonly/js04.jpg


--------------------------------------------------------------------------------
/images-faceonly/js05.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-faceonly/js05.jpg


--------------------------------------------------------------------------------
/images-faceonly/js06.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-faceonly/js06.jpg


--------------------------------------------------------------------------------
/images-faceonly/js07.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-faceonly/js07.jpg


--------------------------------------------------------------------------------
/images-faceonly/js08.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-faceonly/js08.jpg


--------------------------------------------------------------------------------
/images-faceonly/js09.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-faceonly/js09.jpg


--------------------------------------------------------------------------------
/images-faceonly/js10.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-faceonly/js10.jpg


--------------------------------------------------------------------------------
/images-nobkg/js01.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-nobkg/js01.jpg


--------------------------------------------------------------------------------
/images-nobkg/js02.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-nobkg/js02.jpg


--------------------------------------------------------------------------------
/images-nobkg/js03.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-nobkg/js03.jpg


--------------------------------------------------------------------------------
/images-nobkg/js04.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-nobkg/js04.jpg


--------------------------------------------------------------------------------
/images-nobkg/js05.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-nobkg/js05.jpg


--------------------------------------------------------------------------------
/images-nobkg/js06.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-nobkg/js06.jpg


--------------------------------------------------------------------------------
/images-nobkg/js07.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-nobkg/js07.jpg


--------------------------------------------------------------------------------
/images-nobkg/js08.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-nobkg/js08.jpg


--------------------------------------------------------------------------------
/images-nobkg/js09.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-nobkg/js09.jpg


--------------------------------------------------------------------------------
/images-nobkg/js10.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-nobkg/js10.jpg


--------------------------------------------------------------------------------
/images-test-faceonly/test_01_js.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-faceonly/test_01_js.jpg


--------------------------------------------------------------------------------
/images-test-faceonly/test_02_js.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-faceonly/test_02_js.jpg


--------------------------------------------------------------------------------
/images-test-faceonly/test_03_js.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-faceonly/test_03_js.jpg


--------------------------------------------------------------------------------
/images-test-faceonly/test_04_js.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-faceonly/test_04_js.jpg


--------------------------------------------------------------------------------
/images-test-faceonly/test_05_ws.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-faceonly/test_05_ws.jpg


--------------------------------------------------------------------------------
/images-test-faceonly/test_06_kr.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-faceonly/test_06_kr.jpg


--------------------------------------------------------------------------------
/images-test-faceonly/test_07_jc.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-faceonly/test_07_jc.jpg


--------------------------------------------------------------------------------
/images-test-faceonly/test_08_ip.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-faceonly/test_08_ip.jpg


--------------------------------------------------------------------------------
/images-test-nobkg/test_01_js.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-nobkg/test_01_js.jpg


--------------------------------------------------------------------------------
/images-test-nobkg/test_02_js.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-nobkg/test_02_js.jpg


--------------------------------------------------------------------------------
/images-test-nobkg/test_03_js.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-nobkg/test_03_js.jpg


--------------------------------------------------------------------------------
/images-test-nobkg/test_04_js.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-nobkg/test_04_js.jpg


--------------------------------------------------------------------------------
/images-test-nobkg/test_05_ws.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-nobkg/test_05_ws.jpg


--------------------------------------------------------------------------------
/images-test-nobkg/test_06_kr.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-nobkg/test_06_kr.jpg


--------------------------------------------------------------------------------
/images-test-nobkg/test_07_jc.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-nobkg/test_07_jc.jpg


--------------------------------------------------------------------------------
/images-test-nobkg/test_08_ip.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test-nobkg/test_08_ip.jpg


--------------------------------------------------------------------------------
/images-test/test_01_js.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test/test_01_js.jpg


--------------------------------------------------------------------------------
/images-test/test_02_js.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test/test_02_js.jpg


--------------------------------------------------------------------------------
/images-test/test_03_js.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test/test_03_js.jpg


--------------------------------------------------------------------------------
/images-test/test_04_js.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test/test_04_js.jpg


--------------------------------------------------------------------------------
/images-test/test_05_ws.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test/test_05_ws.jpg


--------------------------------------------------------------------------------
/images-test/test_06_kr.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test/test_06_kr.jpg


--------------------------------------------------------------------------------
/images-test/test_07_jc.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test/test_07_jc.jpg


--------------------------------------------------------------------------------
/images-test/test_08_ip.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images-test/test_08_ip.jpg


--------------------------------------------------------------------------------
/images/js01.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images/js01.jpg


--------------------------------------------------------------------------------
/images/js02.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images/js02.jpg


--------------------------------------------------------------------------------
/images/js03.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images/js03.jpg


--------------------------------------------------------------------------------
/images/js04.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images/js04.jpg


--------------------------------------------------------------------------------
/images/js05.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images/js05.jpg


--------------------------------------------------------------------------------
/images/js06.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images/js06.jpg


--------------------------------------------------------------------------------
/images/js07.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images/js07.jpg


--------------------------------------------------------------------------------
/images/js08.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images/js08.jpg


--------------------------------------------------------------------------------
/images/js09.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images/js09.jpg


--------------------------------------------------------------------------------
/images/js10.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ip681/face-recognition-with-correlation-coefficients-and-euclidean-distance/dbb1e3e46d686611e5b35f6ee315d22269e1e41f/images/js10.jpg


--------------------------------------------------------------------------------