├── .github └── workflows │ └── python-app.yml ├── .gitignore ├── 3d-plot.py ├── LICENSE ├── README.md ├── color_gray.py ├── color_sepia.py ├── color_swap.py ├── docs ├── _config.yml ├── color.md ├── face_crop.md ├── face_detection.md ├── filter.md ├── index.md ├── photo_cat.md ├── photo_exif_data_print.md ├── pillow_numpy.md ├── resize.md ├── sift.md └── watershed.md ├── extract_color.py ├── face_crop-all.sh ├── face_crop.py ├── face_crop_raspi.py ├── face_crop_simple-all.sh ├── face_crop_simple.py ├── face_detection.py ├── face_detection_camera.py ├── filter_3by3.py ├── measure_color_average.py ├── photo_cat.py ├── photo_date_print.py ├── photo_exif_date_print.py ├── pillow_numpy_basic.py ├── resize-all.sh ├── resize.py ├── sift.py ├── tests ├── __init__.py └── test_extract_color.py └── watershed.py /.github/workflows/python-app.yml: -------------------------------------------------------------------------------- 1 | # This workflow will install Python dependencies, run tests and lint with a single version of Python 2 | # For more information see: https://help.github.com/actions/language-and-framework-guides/using-python-with-github-actions 3 | 4 | name: Python application 5 | 6 | on: 7 | push: 8 | branches: [ master ] 9 | pull_request: 10 | branches: [ master ] 11 | 12 | jobs: 13 | build: 14 | 15 | runs-on: ubuntu-latest 16 | 17 | steps: 18 | - uses: actions/checkout@v2 19 | - name: Set up Python 3.9 20 | uses: actions/setup-python@v2 21 | with: 22 | python-version: 3.9 23 | - name: Install dependencies 24 | run: | 25 | python -m pip install --upgrade pip 26 | pip install flake8 pytest opencv-python 27 | if [ -f requirements.txt ]; then pip install -r requirements.txt; fi 28 | - name: Lint with flake8 29 | run: | 30 | # stop the build if there are Python syntax errors or undefined names 31 | flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics 32 | # exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide 33 | flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics 34 | - name: Test with pytest 35 | run: | 36 | pytest 37 | -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- 1 | # Byte-compiled / optimized / DLL files 2 | __pycache__/ 3 | *.py[cod] 4 | *$py.class 5 | 6 | # C extensions 7 | *.so 8 | 9 | # Distribution / packaging 10 | .Python 11 | env/ 12 | build/ 13 | develop-eggs/ 14 | dist/ 15 | downloads/ 16 | eggs/ 17 | .eggs/ 18 | lib/ 19 | lib64/ 20 | parts/ 21 | sdist/ 22 | var/ 23 | *.egg-info/ 24 | .installed.cfg 25 | *.egg 26 | 27 | # PyInstaller 28 | # Usually these files are written by a python script from a template 29 | # before PyInstaller builds the exe, so as to inject date/other infos into it. 30 | *.manifest 31 | *.spec 32 | 33 | # Installer logs 34 | pip-log.txt 35 | pip-delete-this-directory.txt 36 | 37 | # Unit test / coverage reports 38 | htmlcov/ 39 | .tox/ 40 | .coverage 41 | .coverage.* 42 | .cache 43 | nosetests.xml 44 | coverage.xml 45 | *,cover 46 | .hypothesis/ 47 | 48 | # Translations 49 | *.mo 50 | *.pot 51 | 52 | # Django stuff: 53 | *.log 54 | 55 | # Sphinx documentation 56 | docs/_build/ 57 | 58 | # PyBuilder 59 | target/ 60 | 61 | #Ipython Notebook 62 | .ipynb_checkpoints 63 | 64 | # For Mac 65 | .DS_Store 66 | -------------------------------------------------------------------------------- /3d-plot.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | from matplotlib import pyplot 3 | from mpl_toolkits.mplot3d import Axes3D 4 | from scipy import genfromtxt 5 | 6 | # read csv file 7 | d = genfromtxt("test.csv", delimiter=",") 8 | 9 | # make graph 10 | fig = pyplot.figure() 11 | ax = Axes3D(fig) 12 | 13 | # setting graph 14 | ax.set_xlabel("Green") 15 | ax.set_ylabel("Blue") 16 | ax.set_zlabel("Red") 17 | 18 | # plot 19 | ax.plot(d[:,1], d[:,0], d[:,2], "o", color="#ff0000", ms=4, mew=0.5) 20 | 21 | pyplot.show() 22 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | The MIT License (MIT) 2 | 3 | Copyright (c) 2016 karaage 4 | 5 | Permission is hereby granted, free of charge, to any person obtaining a copy 6 | of this software and associated documentation files (the "Software"), to deal 7 | in the Software without restriction, including without limitation the rights 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 9 | copies of the Software, and to permit persons to whom the Software is 10 | furnished to do so, subject to the following conditions: 11 | 12 | The above copyright notice and this permission notice shall be included in all 13 | copies or substantial portions of the Software. 14 | 15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 21 | SOFTWARE. 22 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # python-image-processing 2 | Useful and simple image processing python scripts 3 | 4 | image 5 | image 6 | image 7 | image 8 | image 9 | 10 | 11 | # Requirements 12 | - Mac OS X or Linux 13 | - Python3(Recommended) 14 | - Python2 15 | 16 | # Preparation 17 | ## Library 18 | - PIL 19 | - OpenCV 2 20 | - numpy 21 | - matplotlib 22 | - scipy 23 | 24 | Execute following commands for install library: 25 | ```sh 26 | $ pip install pillow 27 | $ pip install opencv 28 | $ pip install numpy 29 | $ pip install matplotlib 30 | $ pip install scipy 31 | ``` 32 | 33 | ## Install(Clone) software 34 | Execute following command: 35 | ```sh 36 | $ git clone https://github.com/karaage0703/python-image-processing.git 37 | ``` 38 | 39 | # Usage and Samples 40 | Go to Github pages. 41 | 42 | [github pages for image processing](https://karaage0703.github.io/python-image-processing) 43 | 44 | # License 45 | This software is released under the MIT License, see LICENSE. 46 | 47 | # Authors 48 | - karaage0703 49 | - [sugyan](https://github.com/sugyan/face-collector) (face_crop.py) 50 | 51 | 52 | -------------------------------------------------------------------------------- /color_gray.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import cv2 3 | import sys 4 | 5 | # extract color function 6 | def color_gray(src): 7 | img_bgr = cv2.split(src) 8 | dst = cv2.merge(((img_bgr[0] + img_bgr[1] + img_bgr[2])/3, (img_bgr[0] + img_bgr[1] + img_bgr[2])/3, (img_bgr[0] + img_bgr[1] + img_bgr[2])/3)) 9 | 10 | return dst 11 | 12 | if __name__ == '__main__': 13 | param = sys.argv 14 | if (len(param) != 2): 15 | print ("Usage: $ python " + param[0] + " sample.jpg") 16 | quit() 17 | # open image file 18 | try: 19 | input_img = cv2.imread(param[1]) 20 | except: 21 | print ('faild to load %s' % param[1]) 22 | quit() 23 | 24 | if input_img is None: 25 | print ('faild to load %s' % param[1]) 26 | quit() 27 | 28 | # making mask using by extract color function 29 | output_img = color_gray(input_img) 30 | 31 | cv2.imwrite("gray_" + param[1], output_img) 32 | -------------------------------------------------------------------------------- /color_sepia.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import cv2 3 | import sys 4 | 5 | # extract color function 6 | def color_sepia(src): 7 | img_bgr = cv2.split(src) 8 | # R=A, G=0.8xA, B=0.55xA 9 | dst = cv2.merge((img_bgr[0] * 0.55 , img_bgr[1] * 0.8, img_bgr[2] * 1.0)) 10 | 11 | return dst 12 | 13 | if __name__ == '__main__': 14 | param = sys.argv 15 | if (len(param) != 2): 16 | print ("Usage: $ python " + param[0] + " sample.jpg") 17 | quit() 18 | # open image file 19 | try: 20 | input_img = cv2.imread(param[1]) 21 | except: 22 | print ('faild to load %s' % param[1]) 23 | quit() 24 | 25 | if input_img is None: 26 | print ('faild to load %s' % param[1]) 27 | quit() 28 | 29 | # making mask using by extract color function 30 | output_img = color_sepia(input_img) 31 | 32 | cv2.imwrite("sepia_" + param[1], output_img) 33 | -------------------------------------------------------------------------------- /color_swap.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import cv2 3 | import sys 4 | 5 | # extract color function 6 | def color_swap(src): 7 | img_bgr = cv2.split(src) 8 | dst = cv2.merge((img_bgr[2], img_bgr[1], img_bgr[0])) # from BGR to RGB 9 | 10 | return dst 11 | 12 | if __name__ == '__main__': 13 | param = sys.argv 14 | if (len(param) != 2): 15 | print ("Usage: $ python " + param[0] + " sample.jpg") 16 | quit() 17 | # open image file 18 | try: 19 | input_img = cv2.imread(param[1]) 20 | except: 21 | print ('faild to load %s' % param[1]) 22 | quit() 23 | 24 | if input_img is None: 25 | print ('faild to load %s' % param[1]) 26 | quit() 27 | 28 | # making mask using by extract color function 29 | output_img = color_swap(input_img) 30 | 31 | cv2.imwrite("swap_" + param[1], output_img) 32 | -------------------------------------------------------------------------------- /docs/_config.yml: -------------------------------------------------------------------------------- 1 | theme: jekyll-theme-cayman -------------------------------------------------------------------------------- /docs/color.md: -------------------------------------------------------------------------------- 1 | [index](./index.md) 2 | 3 | ## color_swap.py 4 | 5 | ### description 6 | color swap from RGB to BGR 7 | 8 | ### preparation 9 | prepare target photo 10 | 11 | ### usage 12 | ```sh 13 | $ python color_swap.py sample.jpg 14 | ``` 15 | 16 | ### original image 17 | image 18 | 19 | 20 | 21 | ### processed image 22 | image 23 | 24 | 25 | ## color_gray.py 26 | 27 | ### description 28 | gray scale converter 29 | 30 | ### preparation 31 | prepare target photo 32 | 33 | ### usage 34 | ```sh 35 | $ python color_gray.py sample.jpg 36 | ``` 37 | 38 | ### original image 39 | image 40 | 41 | 42 | 43 | ### processed image 44 | image 45 | 46 | ## color_sepia.py 47 | 48 | ### description 49 | sepia color converter 50 | 51 | ### preparation 52 | prepare target photo 53 | 54 | ### usage 55 | ```sh 56 | $ python color_sepia.py sample.jpg 57 | ``` 58 | 59 | ### original image 60 | image 61 | 62 | ### processed image 63 | image 64 | 65 | ## extract_color.py 66 | 67 | ### description 68 | extract color 69 | 70 | ### preparation 71 | prepare target photo 72 | 73 | ### usage 74 | ```sh 75 | $ python extract_color.py sample.jpg h_min, h_max, s_th, v_th 76 | ``` 77 | 78 | example: 79 | ```sh 80 | $ python extract_color.py 0002.jpg 10 40 10 50 81 | ``` 82 | 83 | ### original image 84 | image 85 | 86 | ### processed image 87 | image 88 | 89 | ## References 90 | - Interface 2017/05 91 | - http://venuschjp.blogspot.jp/2015/02/pythonopencv.html 92 | - https://www.blog.umentu.work/python-opencv3で画素のrgb値を入れ替える/ 93 | -------------------------------------------------------------------------------- /docs/face_crop.md: -------------------------------------------------------------------------------- 1 | ## face_crop.py 2 | [index](./index.md) 3 | 4 | ### preparation 5 | 6 | ### usage 7 | Execute following command: 8 | ```sh 9 | $ python face_crop.py sample.jpg 10 | ``` 11 | 12 | If you want convert all files in same directory with this script, Execute following command: 13 | ```sh 14 | $ ./face_crop-all.sh 15 | ``` 16 | 17 | ### original image 18 | image 19 | 20 | ### processed image 21 | image 22 | 23 | 24 | # Reference 25 | - http://memo.sugyan.com/entry/20151203/1449137219 26 | -------------------------------------------------------------------------------- /docs/face_detection.md: -------------------------------------------------------------------------------- 1 | ## face_detection.py 2 | [index](./index.md) 3 | 4 | ### preparation 5 | Search `haarcascade_frontalface_alt.xml`: 6 | ```sh 7 | $ sudo find / -name "haarcascade_frontalface_alt.xml" 8 | ``` 9 | 10 | and copy file to this repository directory 11 | 12 | If you cannot find file, please download from web 13 | 14 | ### usage 15 | ```sh 16 | $ python face_detection.py sample.jpg 17 | ``` 18 | 19 | ### original image 20 | image 21 | 22 | ### processed image 23 | image 24 | 25 | 26 | ## face_detection_camera.py 27 | 28 | ### caution 29 | Camera is needed 30 | 31 | ### preparation 32 | Search `haarcascade_frontalface_alt.xml`: 33 | ```sh 34 | $ sudo find / -name "haarcascade_frontalface_alt.xml" 35 | ``` 36 | and copy file to this repository directory 37 | 38 | If you cannot find file, please download from web 39 | 40 | ### command 41 | ```sh 42 | $ python face_detection_camera.py 43 | ``` 44 | 45 | # Reference 46 | -------------------------------------------------------------------------------- /docs/filter.md: -------------------------------------------------------------------------------- 1 | [index](./index.md) 2 | 3 | ## filter_3by3.py 4 | 5 | ### description 6 | Image processing by using 3x3 filter 7 | This script is laplacian filter 8 | 9 | ### preparation 10 | prepare target photo 11 | 12 | ### usage 13 | ```sh 14 | $ python filter_3by3.py sample.jpg 15 | ``` 16 | 17 | ### original image 18 | image 19 | 20 | 21 | 22 | ### processed image 23 | image 24 | 25 | ## References 26 | - Interface 2017/05 27 | -------------------------------------------------------------------------------- /docs/index.md: -------------------------------------------------------------------------------- 1 | [home](https://github.com/karaage0703/python-image-processing) 2 | 3 | # python-image-processing-sample 4 | 5 | - [resize](./resize.md) 6 | - [photo synthesis](./photo_cat.md) 7 | - [data print to photo by using exif data](./photo_exif_data_print.md) 8 | - [face detection](./face_detection.md) 9 | - [face crop](./face_crop.md) 10 | - [sift](./sift.md) 11 | - [pillow and numpy basic](./pillow_numpy.md) 12 | - [color](./color.md) 13 | - [filter](./filter.md) 14 | - [watershed](./watershed.md) 15 | 16 | -------------------------------------------------------------------------------- /docs/photo_cat.md: -------------------------------------------------------------------------------- 1 | [index](./index.md) 2 | 3 | ## photo-cat.py 4 | 5 | ### description 6 | Synth one photo from four photos. 7 | 8 | ### preparation 9 | prepare 4 image files named `image-1.jpg` `image-2.jpg` `image-3.jpg` `image-4.jpg` 10 | 11 | ### command 12 | ```sh 13 | $ python photo_cat.py 14 | ``` 15 | 16 | ### original image 17 | image 18 | 19 | image 20 | 21 | image 22 | 23 | image 24 | 25 | ### processed image 26 | image 27 | 28 | ## Reference 29 | -------------------------------------------------------------------------------- /docs/photo_exif_data_print.md: -------------------------------------------------------------------------------- 1 | ## photo-exif-date-print.py 2 | [index](./index.md) 3 | 4 | ### description 5 | Print date to photo by using exif data. 6 | 7 | ### preparation 8 | Search font file place of your PC: 9 | ```sh 10 | $ sudo find / -name "Arial Black.ttf" 11 | ``` 12 | 13 | 　Mac example 14 | ``` 15 | /Library/Fonts/Arial Black.ttf 16 | ``` 17 | 18 | 　copy font file 19 | ```sh 20 | $ cp /Library/Fonts/Arial Black.ttf ./ 21 | ``` 22 | 23 | ### usage 24 | ```sh 25 | $ python photo_exif_date_print.py sample.jpg 26 | ``` 27 | 28 | if you want to process many files. make below script named `photo-exif-date-print.sh` 29 | ```sh 30 | #!/bin/bash 31 | for f in *.jpg 32 | do 33 | python photo-exif-date-print.py $f 34 | done 35 | ``` 36 | 37 | then execute following commands: 38 | ```sh 39 | $ chmod 755 photo-exif-date-print.sh 40 | $ ./photo-exif-date-print.sh 41 | ``` 42 | 43 | ### original image 44 | image 45 | 46 | ### processed image 47 | image 48 | 49 | ## Reference 50 | -------------------------------------------------------------------------------- /docs/pillow_numpy.md: -------------------------------------------------------------------------------- 1 | [index](./index.md) 2 | 3 | ## pillow_numpy_basic.py 4 | 5 | ### description 6 | image processing basic sample (this script no processing) 7 | 8 | ### preparation 9 | prepare target photo 10 | 11 | ### usage 12 | ```sh 13 | $ python pillow_numpy_basic.py sample.jpg 14 | ``` 15 | 16 | ### original image 17 | any image is ok 18 | 19 | ### processed image 20 | same with original 21 | 22 | ## References 23 | - http://qiita.com/uosansatox/items/4fa34e1d8d95d8783536 24 | - http://d.hatena.ne.jp/white_wheels/20100322/p1 25 | -------------------------------------------------------------------------------- /docs/resize.md: -------------------------------------------------------------------------------- 1 | [index](./index.md) 2 | 3 | ## resize.py 4 | 5 | ### description 6 | resize photo 7 | 8 | ### preparation 9 | prepare target photo 10 | 11 | ### usage 12 | ```sh 13 | $ python resize.py sample.jpg wide_ratio height_ratio 14 | ``` 15 | 16 | example: 17 | ```sh 18 | $ python resize.py sample.jpg 10 10 19 | ``` 20 | 21 | ### original image 22 | image 23 | 24 | ### processed image 25 | image 26 | 27 | ## resize-all.sh 28 | 29 | ### description 30 | resize all photo in directory 31 | 32 | ### preparation 33 | prepare target photos in directory 34 | 35 | ### usage 36 | ```sh 37 | $ ./resize-all.sh 38 | ``` 39 | 40 | ## References 41 | - http://tatabox.hatenablog.com/entry/2013/07/15/164015 42 | -------------------------------------------------------------------------------- /docs/sift.md: -------------------------------------------------------------------------------- 1 | [index](./index.md) 2 | 3 | ## shift.py 4 | 5 | ### description 6 | Test SIFT 7 | 8 | ### preparation 9 | prepare 2 image files 10 | 11 | ### command 12 | ```sh 13 | $ python shift.py sample.jpg sample2.jpg 14 | ``` 15 | 16 | ### original image 17 | image 18 | 19 | image 20 | 21 | ### processed image 22 | image 23 | 24 | ## References 25 | - http://toaru-gi.hatenablog.com/entry/2014/10/27/040843 26 | -------------------------------------------------------------------------------- /docs/watershed.md: -------------------------------------------------------------------------------- 1 | [index](./index.md) 2 | 3 | ## watershed.py 4 | 5 | ### description 6 | Image segmented by watershed algorithm 7 | 8 | ### preparation 9 | Download sample photo. 10 | Execute following command: 11 | 12 | ```sh 13 | $ wget https://docs.opencv.org/3.1.0/water_coins.jpg 14 | ``` 15 | 16 | ### usage 17 | ```sh 18 | $ python watershed.py water_coins.jpg 19 | ``` 20 | 21 | ### original image 22 | image 23 | 24 | ### processed image 25 | #### segmented image 26 | image 27 | 28 | #### mask image 29 | image 30 | 31 | ## References 32 | - https://qiita.com/ysdyt/items/5972c9520acf6a094d90 33 | -------------------------------------------------------------------------------- /extract_color.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import cv2 3 | import sys 4 | import numpy as np 5 | 6 | # extract color function 7 | def extract_color( src, h_th_low, h_th_up, s_th, v_th ): 8 | hsv = cv2.cvtColor(src, cv2.COLOR_BGR2HSV) 9 | h, s, v = cv2.split(hsv) 10 | if h_th_low > h_th_up: 11 | ret, h_dst_1 = cv2.threshold(h, h_th_low, 255, cv2.THRESH_BINARY) 12 | ret, h_dst_2 = cv2.threshold(h, h_th_up, 255, cv2.THRESH_BINARY_INV) 13 | 14 | dst = cv2.bitwise_or(h_dst_1, h_dst_2) 15 | else: 16 | ret, dst = cv2.threshold(h, h_th_low, 255, cv2.THRESH_TOZERO) 17 | ret, dst = cv2.threshold(dst, h_th_up, 255, cv2.THRESH_TOZERO_INV) 18 | ret, dst = cv2.threshold(dst, 0, 255, cv2.THRESH_BINARY) 19 | 20 | ret, s_dst = cv2.threshold(s, s_th, 255, cv2.THRESH_BINARY) 21 | ret, v_dst = cv2.threshold(v, v_th, 255, cv2.THRESH_BINARY) 22 | dst = cv2.bitwise_and(dst, s_dst) 23 | dst = cv2.bitwise_and(dst, v_dst) 24 | return dst 25 | 26 | if __name__ == '__main__': 27 | param = sys.argv 28 | if (len(param) != 6): 29 | print ("Usage: $ python " + param[0] + " sample.jpg h_min, h_max, s_th, v_th") 30 | quit() 31 | # open image file 32 | try: 33 | input_img = cv2.imread(param[1]) 34 | except: 35 | print ('faild to load %s' % param[1]) 36 | quit() 37 | 38 | if input_img is None: 39 | print ('faild to load %s' % param[1]) 40 | quit() 41 | 42 | # parameter setting 43 | h_min = int(param[2]) 44 | h_max = int(param[3]) 45 | s_th = int(param[4]) 46 | v_th = int(param[5]) 47 | # making mask using by extract color function 48 | msk_img = extract_color(input_img, h_min, h_max, s_th, v_th) 49 | # mask processing 50 | output_img = cv2.bitwise_and(input_img, input_img, mask = msk_img) 51 | 52 | cv2.imwrite("extract_" + param[1], output_img) 53 | -------------------------------------------------------------------------------- /face_crop-all.sh: -------------------------------------------------------------------------------- 1 | #!/bin/bash 2 | 3 | echo face crop processing... 4 | for f in *.JPG 5 | do 6 | python face_crop.py $f 7 | done 8 | 9 | for f in *.jpg 10 | do 11 | python face_crop.py $f 12 | done 13 | 14 | for f in *.JPEG 15 | do 16 | python face_crop.py $f 17 | done 18 | 19 | for f in *.jpeg 20 | do 21 | python face_crop.py $f 22 | done 23 | 24 | echo done 25 | -------------------------------------------------------------------------------- /face_crop.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import cv2 3 | import numpy as np 4 | import sys 5 | import math 6 | from math import sin, cos 7 | from os import path 8 | 9 | cascades_dir = path.normpath(path.join(cv2.__file__, '..', '..', '..', '..', 'share', 'OpenCV', 'haarcascades')) 10 | max_size = 720 11 | 12 | def detect(img, filename): 13 | cascade_f = cv2.CascadeClassifier(path.join(cascades_dir, 'haarcascade_frontalface_alt2.xml')) 14 | cascade_e = cv2.CascadeClassifier(path.join(cascades_dir, 'haarcascade_eye.xml')) 15 | # resize if learch image 16 | rows, cols, _ = img.shape 17 | if max(rows, cols) > max_size: 18 | l = max(rows, cols) 19 | img = cv2.resize(img, (cols * max_size / l, rows * max_size / l)) 20 | rows, cols, _ = img.shape 21 | # create gray image for rotate 22 | gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) 23 | hypot = int(math.ceil(math.hypot(rows, cols))) 24 | frame = np.zeros((hypot, hypot), np.uint8) 25 | frame[(hypot - rows) * 0.5:(hypot + rows) * 0.5, (hypot - cols) * 0.5:(hypot + cols) * 0.5] = gray 26 | 27 | def translate(coord, deg): 28 | x, y = coord 29 | rad = math.radians(deg) 30 | return { 31 | 'x': ( cos(rad) * x + sin(rad) * y - hypot * 0.5 * cos(rad) - hypot * 0.5 * sin(rad) + hypot * 0.5 - (hypot - cols) * 0.5) / float(cols) * 100.0, 32 | 'y': (- sin(rad) * x + cos(rad) * y + hypot * 0.5 * sin(rad) - hypot * 0.5 * cos(rad) + hypot * 0.5 - (hypot - rows) * 0.5) / float(rows) * 100.0, 33 | } 34 | 35 | # filename numbering 36 | numb = 0 37 | for deg in range(-48, 49, 6): 38 | M = cv2.getRotationMatrix2D((hypot * 0.5, hypot * 0.5), deg, 1.0) 39 | rotated = cv2.warpAffine(frame, M, (hypot, hypot)) 40 | faces = cascade_f.detectMultiScale(rotated, 1.08, 2) 41 | print(deg, len(faces)) 42 | for face in faces: 43 | x, y, w, h = face 44 | # eyes in face? 45 | y_offset = int(h * 0.1) 46 | roi = rotated[y + y_offset: y + h, x: x + w] 47 | eyes = cascade_e.detectMultiScale(roi, 1.05) 48 | eyes = filter(lambda e: (e[0] > w / 2 or e[0] + e[2] < w / 2) and e[1] + e[3] < h / 2, eyes) 49 | if len(eyes) == 2 and abs(eyes[0][0] - eyes[1][0]) > w / 4: 50 | score = math.atan2(abs(eyes[1][1] - eyes[0][1]), abs(eyes[1][0] - eyes[0][0])) 51 | if eyes[0][1] == eyes[1][1]: 52 | score = 0.0 53 | if score < 0.15: 54 | print("face_pos=" + str(face)) 55 | print("score=" + str(score)) 56 | print("numb=" + str(numb)) 57 | # cv2.rectangle(rotated, (x ,y), (x+w, x+h), (255,255,255), thickness=2) 58 | # crop image 59 | output_img = rotated[y:y+h, x:x+h] 60 | # output file 61 | cv2.imwrite(str("{0:02d}".format(numb)) + "_face_" + filename, output_img) 62 | numb += 1 63 | 64 | 65 | if __name__ == '__main__': 66 | param = sys.argv 67 | if (len(param) != 2): 68 | print ("Usage: $ python " + param[0] + " sample.jpg") 69 | quit() 70 | 71 | # open image file 72 | try: 73 | input_img = cv2.imread(param[1]) 74 | except: 75 | print ('faild to load %s' % param[1]) 76 | quit() 77 | 78 | if input_img is None: 79 | print ('faild to load %s' % param[1]) 80 | quit() 81 | 82 | output_img = detect(input_img, param[1]) 83 | -------------------------------------------------------------------------------- /face_crop_raspi.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import cv2 3 | import sys 4 | from os import path 5 | import math 6 | from math import sin, cos 7 | 8 | # cascades_dir = path.normpath(path.join(cv2.__file__, '..', '..', '..', '..', 'share', 'OpenCV', 'haarcascades')) 9 | cascades_dir ='/usr/share/opencv/haarcascades' 10 | 11 | color = (255, 255, 255) # color of rectangle for face detection 12 | 13 | def face_detect(file): 14 | image = cv2.imread(file) 15 | image_gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY) 16 | cascade_f = cv2.CascadeClassifier(path.join(cascades_dir, 'haarcascade_frontalface_alt2.xml')) 17 | cascade_e = cv2.CascadeClassifier(path.join(cascades_dir, 'haarcascade_eye.xml')) 18 | 19 | # facerect = cascade_f.detectMultiScale(image_gray, scaleFactor=1.08, minNeighbors=1, minSize=(50, 50)) 20 | facerect = cascade_f.detectMultiScale(image_gray, scaleFactor=1.08, minNeighbors=1, minSize=(200, 200)) 21 | 22 | print("face rectangle") 23 | print(facerect) 24 | 25 | # test_image = image 26 | 27 | if len(facerect) > 0: 28 | # filename numbering 29 | numb = 0 30 | for rect in facerect: 31 | x, y, w, h = rect 32 | # eyes in face? 33 | roi = image_gray[y: y + h, x: x + w] 34 | # cv2.rectangle(test_image,(x,y),(x+w,y+h),(255,255,255),2) 35 | eyes = cascade_e.detectMultiScale(roi, scaleFactor=1.05, minSize=(20,20)) 36 | # eyes = filter(lambda e: (e[0] > w / 2 or e[0] + e[2] < w / 2) and e[1] + e[3] < h / 2, eyes) 37 | # for (ex,ey,ew,eh) in eyes: 38 | # print("rectangle") 39 | # cv2.rectangle(test_image,(x+ex,y+ey),(x+ex+ew,y+ey+eh),(0,255,0),2) 40 | if len(eyes) > 1: 41 | # if len(eyes) == 2 and abs(eyes[0][0] - eyes[1][0]) > w / 4: 42 | image_face = image[y:y+h, x:x+h] 43 | cv2.imwrite(str("{0:02d}".format(numb)) + "_face_" + file, image_face) 44 | numb += 1 45 | 46 | # cv2.imwrite("00debug_"+file, test_image) 47 | 48 | if __name__ == '__main__': 49 | param = sys.argv 50 | if (len(param) != 2): 51 | print ("Usage: $ python " + param[0] + " sample.jpg") 52 | quit() 53 | 54 | face_detect(param[1]) 55 | -------------------------------------------------------------------------------- /face_crop_simple-all.sh: -------------------------------------------------------------------------------- 1 | #!/bin/bash 2 | 3 | cmd=face_crop_simple.py 4 | echo face crop processing... 5 | for f in *.jpg *.jpeg *.JPG *.JPEG *.png *.PNG 6 | do 7 | if [ -f $f ]; then 8 | echo python $cmd $f 9 | python $cmd $f 10 | fi 11 | done 12 | 13 | echo done 14 | -------------------------------------------------------------------------------- /face_crop_simple.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import cv2 3 | import sys 4 | import os 5 | from os import path 6 | import math 7 | from math import sin, cos 8 | 9 | cascades_dir = path.normpath(path.join(cv2.__file__, '..', '..', '..', '..', 'share', 'OpenCV', 'haarcascades')) 10 | 11 | color = (255, 255, 255) # color of rectangle for face detection 12 | 13 | def face_detect(file): 14 | image = cv2.imread(file) 15 | image_gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY) 16 | cascade_f = cv2.CascadeClassifier(path.join(cascades_dir, 'haarcascade_frontalface_alt2.xml')) 17 | cascade_e = cv2.CascadeClassifier(path.join(cascades_dir, 'haarcascade_eye.xml')) 18 | 19 | facerect = cascade_f.detectMultiScale(image_gray, scaleFactor=1.08, minNeighbors=1, minSize=(50, 50)) 20 | 21 | # print("face rectangle") 22 | # print(facerect) 23 | if not os.path.exists("face_images"): 24 | os.mkdir("face_images") 25 | 26 | base = os.path.splitext(os.path.basename(sys.argv[1]))[0] + "_" 27 | 28 | if len(facerect) > 0: 29 | # filename numbering 30 | numb = 0 31 | for rect in facerect: 32 | x, y, w, h = rect 33 | # eyes in face? 34 | y_offset = int(h * 0.1) 35 | eye_area = image_gray[y + y_offset: y + h, x: x + w] 36 | eyes = cascade_e.detectMultiScale(eye_area, 1.05) 37 | eyes = filter(lambda e: (e[0] > w / 2 or e[0] + e[2] < w / 2) and e[1] + e[3] < h / 2, eyes) 38 | # print(len(eyes)) 39 | if len(eyes) > 0: 40 | image_face = image[y:y+h, x:x+h] 41 | cv2.imwrite("face_images/" + base + str("{0:02d}".format(numb)) + ".jpg", image_face) 42 | numb += 1 43 | 44 | if __name__ == '__main__': 45 | param = sys.argv 46 | if (len(param) != 2): 47 | print ("Usage: $ python " + param[0] + " sample.jpg") 48 | quit() 49 | 50 | face_detect(param[1]) 51 | -------------------------------------------------------------------------------- /face_detection.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import cv2 3 | import sys 4 | cascade_path = "./haarcascade_frontalface_alt.xml" 5 | 6 | color = (255, 255, 255) # color of rectangle for face detection 7 | 8 | def face_detect(file): 9 | image = cv2.imread(file) 10 | image_gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY) 11 | cascade = cv2.CascadeClassifier(cascade_path) 12 | facerect = cascade.detectMultiScale(image_gray, scaleFactor=1.1, minNeighbors=1, minSize=(1, 1)) 13 | 14 | print("face rectangle") 15 | print(facerect) 16 | 17 | if len(facerect) > 0: 18 | for rect in facerect: 19 | cv2.rectangle(image, tuple(rect[0:2]),tuple(rect[0:2]+rect[2:4]), color, thickness=2) 20 | 21 | return image 22 | 23 | if __name__ == '__main__': 24 | param = sys.argv 25 | if (len(param) != 2): 26 | print ("Usage: $ python " + param[0] + " sample.jpg") 27 | quit() 28 | 29 | output_img = face_detect(param[1]) 30 | cv2.imwrite('facedetect_' + param[1], output_img) 31 | -------------------------------------------------------------------------------- /face_detection_camera.py: -------------------------------------------------------------------------------- 1 | #!/usr/bin/env python 2 | # -*- coding: utf-8 -*- 3 | import cv2 4 | cascade_path = "./haarcascade_frontalface_alt.xml" 5 | color = (255, 255, 255) # color of rectangle for face detection 6 | 7 | cam = cv2.VideoCapture(0) 8 | count=0 9 | 10 | while True: 11 | ret, capture = cam.read() 12 | if not ret: 13 | print('error') 14 | break 15 | count += 1 16 | if count > 1: 17 | image = capture.copy() 18 | image_gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY) 19 | cascade = cv2.CascadeClassifier(cascade_path) 20 | facerect = cascade.detectMultiScale(image_gray, scaleFactor=1.1, minNeighbors=1, minSize=(1, 1)) 21 | 22 | if len(facerect) > 0: 23 | for rect in facerect: 24 | cv2.rectangle(image, tuple(rect[0:2]),tuple(rect[0:2]+rect[2:4]), color, thickness=2) 25 | 26 | count=0 27 | cv2.imshow('face detector', image) 28 | 29 | cam.release() 30 | cv2.destroyAllWindows() 31 | -------------------------------------------------------------------------------- /filter_3by3.py: -------------------------------------------------------------------------------- 1 | from PIL import Image 2 | import numpy as np 3 | import sys 4 | 5 | # filter = [0, 0, 0, 0, 1, 0, 0, 0, 0] 6 | # filter = [0, 1, 0, 1, -4, 1, 0, 1, 0] 7 | # filter = [0, 1, 0, 0, -2, 0, 0, 1, 0] 8 | # filter = [0, 0, 0, 1, -2, 1, 0, 0, 0] 9 | filter = [0, 1, 0, 1, -4, 1, 0, 1, 0] 10 | # filter = [1.0/9.0, 1.0/9.0, 1.0/9.0, 1.0/9.0, 1.0/9.0, 1.0/9.0, 1.0/9.0, 1.0/9.0, 1.0/9.0] 11 | 12 | def image_process(src): 13 | width, height = src.size 14 | dst = Image.new('RGB', (width, height)) 15 | 16 | img_pixels = np.array([[src.getpixel((x,y)) for y in range(height)] for x in range(width)]) 17 | color = np.zeros((len(filter), 3)) 18 | 19 | for y in range(1, height-1): 20 | for x in range(1, width-1): 21 | color[0] = img_pixels[x-1][y-1] 22 | color[1] = img_pixels[x-1][y] 23 | color[2] = img_pixels[x-1][y+1] 24 | color[3] = img_pixels[x][y-1] 25 | color[4] = img_pixels[x][y] 26 | color[5] = img_pixels[x][y+1] 27 | color[6] = img_pixels[x+1][y-1] 28 | color[7] = img_pixels[x+1][y] 29 | color[8] = img_pixels[x+1][y+1] 30 | 31 | sum_color = np.zeros(3) 32 | for num in range(len(filter)): 33 | sum_color += color[num] * filter[num] 34 | 35 | r,g,b = map(int, (sum_color)) 36 | r = min([r, 255]) 37 | r = max([r, 0]) 38 | g = min([g, 255]) 39 | g = max([g, 0]) 40 | b = min([b, 255]) 41 | b = max([b, 0]) 42 | 43 | dst.putpixel((x,y), (r,g,b)) 44 | 45 | return dst 46 | 47 | if __name__ == '__main__': 48 | param = sys.argv 49 | if (len(param) != 2): 50 | print("Usage: $ python " + param[0] + " sample.jpg") 51 | quit() 52 | 53 | # open image file 54 | try: 55 | input_img = Image.open(param[1]) 56 | except: 57 | print('faild to load %s' % param[1]) 58 | quit() 59 | 60 | if input_img is None: 61 | print('faild to load %s' % param[1]) 62 | quit() 63 | 64 | output_img = image_process(input_img) 65 | output_img.save("filtered_" + param[1]) 66 | output_img.show() 67 | -------------------------------------------------------------------------------- /measure_color_average.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import cv2 3 | import sys 4 | import numpy as np 5 | 6 | if __name__ == '__main__': 7 | param = sys.argv 8 | if (len(param) != 3): 9 | print("Usage: $ python " + param[0] + " sample.jpg" "rgb or hsv") 10 | quit() 11 | 12 | # open image file 13 | bgr_img = cv2.imread(param[1]) 14 | hsv_img = cv2.cvtColor(bgr_img, cv2.COLOR_BGR2HSV) 15 | 16 | average_bgr = [0,0,0] 17 | average_hsv = [0,0,0] 18 | 19 | # measure average of RGB color 20 | for i in range(3): 21 | extract_img = bgr_img[:,:,i] 22 | extract_img = extract_img[extract_img>0] 23 | average_bgr[i] = np.average(extract_img) 24 | 25 | for i in range(3): 26 | extract_img = hsv_img[:,:,i] 27 | extract_img = extract_img[extract_img>0] 28 | average_hsv[i] = np.average(extract_img) 29 | 30 | if param[2] == "rgb": 31 | # save RGB format(from BGR) 32 | print(str(average_bgr[2])+","+str(average_bgr[1])+","+str(average_bgr[0])) 33 | elif param[2] == "hsv": 34 | # save HSV format 35 | print(str(average_hsv[0])+","+str(average_hsv[1])+","+str(average_hsv[2])) 36 | else: 37 | print("Option is wrong. please select rgb or hsv") 38 | -------------------------------------------------------------------------------- /photo_cat.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import cv2 3 | 4 | img1 = cv2.imread('image-1.jpg') 5 | img2 = cv2.imread('image-2.jpg') 6 | img3 = cv2.imread('image-3.jpg') 7 | img4 = cv2.imread('image-4.jpg') 8 | 9 | img5 = cv2.vconcat([img1, img2]) 10 | img6 = cv2.vconcat([img3, img4]) 11 | img7 = cv2.hconcat([img5, img6]) 12 | cv2.imwrite('output.jpg', img7) 13 | -------------------------------------------------------------------------------- /photo_date_print.py: -------------------------------------------------------------------------------- 1 | import cv2 2 | import numpy as np 3 | from PIL import Image, ImageDraw, ImageFont 4 | import sys 5 | 6 | font_color = (255, 255, 255) 7 | 8 | 9 | def print_date(file, date): 10 | base_img_cv2 = cv2.imread(file) 11 | 12 | base_img = Image.open(file).convert('RGBA') 13 | txt = Image.new('RGB', base_img.size, (0, 0, 0)) 14 | draw = ImageDraw.Draw(txt) 15 | fnt = ImageFont.truetype('./NotoSansJP-Regular.ttf', 16 | size=(int)((base_img.size[0]+base_img.size[1])/100)) 17 | 18 | bbox = draw.textbbox((0, 0), date, font=fnt) 19 | text_width, text_height = bbox[2] - bbox[0], bbox[3] - bbox[1] 20 | 21 | draw.text(((base_img.size[0]*0.95 - text_width), 22 | (base_img.size[1]*0.95 - text_height)), 23 | date, font=fnt, fill=font_color) 24 | 25 | txt_array = np.array(txt) 26 | 27 | output_img = cv2.addWeighted(base_img_cv2, 1.0, txt_array, 1.0, 0) 28 | return output_img 29 | 30 | 31 | if __name__ == '__main__': 32 | param = sys.argv 33 | if (len(param) != 3): 34 | print("Usage: $ python " + param[0] + " ") 35 | quit() 36 | 37 | output_img = print_date(param[1], param[2]) 38 | cv2.imwrite(param[2].replace(':', '_').replace(' ', '_') + "_" + param[1], output_img) 39 | -------------------------------------------------------------------------------- /photo_exif_date_print.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import cv2 3 | import sys 4 | import numpy as np 5 | from PIL import Image, ImageDraw, ImageFont 6 | from PIL.ExifTags import TAGS, GPSTAGS 7 | 8 | font_color = (255, 255, 255) 9 | 10 | def get_exif(file,field): 11 | img = Image.open(file) 12 | exif = img._getexif() 13 | 14 | exif_data = [] 15 | for id, value in exif.items(): 16 | if TAGS.get(id) == field: 17 | tag = TAGS.get(id, id),value 18 | exif_data.extend(tag) 19 | 20 | return exif_data 21 | 22 | 23 | def get_exif_of_image(file): 24 | im = Image.open(file) 25 | 26 | try: 27 | exif = im._getexif() 28 | except AttributeError: 29 | return {} 30 | 31 | exif_table = {} 32 | for tag_id, value in exif.items(): 33 | tag = TAGS.get(tag_id, tag_id) 34 | exif_table[tag] = value 35 | 36 | return exif_table 37 | 38 | 39 | def get_date_of_image(file): 40 | exif_table = get_exif_of_image(file) 41 | return exif_table.get("DateTimeOriginal") 42 | 43 | 44 | def put_date(file, date): 45 | base_img_cv2 = cv2.imread(file) 46 | 47 | base_img = Image.open(file).convert('RGBA') 48 | txt = Image.new('RGB', base_img.size, (0, 0, 0)) 49 | draw = ImageDraw.Draw(txt) 50 | fnt = ImageFont.truetype('./Arial Black.ttf', size=(int)((base_img.size[0]+base_img.size[1])/100)) 51 | 52 | textw, texth = draw.textsize(date, font=fnt) 53 | 54 | draw.text(((base_img.size[0]*0.95 - textw) , (base_img.size[1]*0.95 - texth)), 55 | date, font=fnt, fill=font_color) 56 | 57 | txt_array = np.array(txt) 58 | 59 | output_img = cv2.addWeighted(base_img_cv2, 1.0, txt_array, 1.0, 0) 60 | return output_img 61 | 62 | 63 | if __name__ == '__main__': 64 | param = sys.argv 65 | if (len(param) != 2): 66 | print ("Usage: $ python " + param[0] + " sample.jpg") 67 | quit() 68 | 69 | try: 70 | date = get_date_of_image(param[1]) 71 | output_img = put_date(param[1], date) 72 | cv2.imwrite(date.replace(':','_').replace(' ','_') + "_" + param[1], output_img) 73 | except: 74 | base_img_cv2 = cv2.imread(param[1]) 75 | cv2.imwrite("nodate_" + param[1], base_img_cv2) 76 | -------------------------------------------------------------------------------- /pillow_numpy_basic.py: -------------------------------------------------------------------------------- 1 | from PIL import Image 2 | import numpy as np 3 | import sys 4 | 5 | def image_process(src): 6 | width, height = src.size 7 | dst = Image.new('RGB', (width, height)) 8 | 9 | img_pixels = np.array([[src.getpixel((x,y)) for y in range(height)] for x in range(width)]) 10 | 11 | for y in range(height): 12 | for x in range(width): 13 | r,g,b = img_pixels[x][y] 14 | dst.putpixel((x,y), (r,g,b)) 15 | 16 | return dst 17 | 18 | if __name__ == '__main__': 19 | param = sys.argv 20 | if (len(param) != 2): 21 | print ("Usage: $ python " + param[0] + " sample.jpg") 22 | quit() 23 | 24 | # open image file 25 | try: 26 | input_img = Image.open(param[1]) 27 | except: 28 | print ('faild to load %s' % param[1]) 29 | quit() 30 | 31 | if input_img is None: 32 | print ('faild to load %s' % param[1]) 33 | quit() 34 | 35 | output_img = image_process(input_img) 36 | output_img.save("process_" + param[1]) 37 | output_img.show() 38 | -------------------------------------------------------------------------------- /resize-all.sh: -------------------------------------------------------------------------------- 1 | #!/bin/bash 2 | 3 | WIDTH_RATIO=10 4 | HEIGHT_RATIO=10 5 | 6 | echo resize processing... 7 | for f in *.JPG 8 | do 9 | echo python resize.py $f $WIDTH_RATIO $HEIGHT_RATIO 10 | python resize.py $f $WIDTH_RATIO $HEIGHT_RATIO 11 | done 12 | echo done 13 | -------------------------------------------------------------------------------- /resize.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import cv2 3 | import sys 4 | 5 | 6 | def resize(src, w_ratio, h_ratio): 7 | """Resize image 8 | 9 | Args: 10 | src (numpy.ndarray): Image 11 | w_ratio (int): Width ratio 12 | h_ratio (int): Height ratio 13 | 14 | Returns: 15 | numpy.ndarray: Image 16 | """ 17 | 18 | height = src.shape[0] 19 | width = src.shape[1] 20 | dst = cv2.resize(src, ((int)(width/100*w_ratio), (int)(height/100*h_ratio))) 21 | return dst 22 | 23 | 24 | if __name__ == '__main__': 25 | param = sys.argv 26 | if(len(param) != 4): 27 | print("Usage: $ python " + param[0] + " sample.jpg wide_ratio height_ratio") 28 | quit() 29 | 30 | # open image file 31 | try: 32 | input_img = cv2.imread(param[1]) 33 | except Exception as e: 34 | print(e) 35 | quit() 36 | 37 | if input_img is None: 38 | print('faild to load %s' % param[1]) 39 | quit() 40 | 41 | w_ratio = int(param[2]) 42 | h_ratio = int(param[3]) 43 | 44 | output_img = resize(input_img, w_ratio, h_ratio) 45 | cv2.imwrite(param[1], output_img) 46 | -------------------------------------------------------------------------------- /sift.py: -------------------------------------------------------------------------------- 1 | import cv2 2 | import sys 3 | import numpy as np 4 | 5 | def drawMatches(img1, kp1, img2, kp2, matches): 6 | rows1 = img1.shape[0] 7 | cols1 = img1.shape[1] 8 | rows2 = img2.shape[0] 9 | cols2 = img2.shape[1] 10 | 11 | out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8') 12 | 13 | out[:rows1,:cols1,:] = np.dstack([img1, img1, img1]) 14 | 15 | out[:rows2,cols1:cols1+cols2,:] = np.dstack([img2, img2, img2]) 16 | 17 | for mat in matches: 18 | 19 | img1_idx = mat.queryIdx 20 | img2_idx = mat.trainIdx 21 | 22 | (x1,y1) = kp1[img1_idx].pt 23 | (x2,y2) = kp2[img2_idx].pt 24 | 25 | cv2.circle(out, (int(x1),int(y1)), 4, (255, 0, 0), 1) 26 | cv2.circle(out, (int(x2)+cols1,int(y2)), 4, (255, 0, 0), 1) 27 | 28 | cv2.line(out, (int(x1),int(y1)), (int(x2)+cols1,int(y2)), (255, 0, 0), 1) 29 | 30 | cv2.imshow('Matched Features', out) 31 | cv2.waitKey(0) 32 | cv2.destroyAllWindows() 33 | 34 | if __name__ == '__main__': 35 | param = sys.argv 36 | if (len(param) != 3): 37 | print ("Usage: $ python " + param[0] + " sample1.jpg sample2.jgp") 38 | quit() 39 | 40 | # open image file 41 | try: 42 | input_img1 = cv2.imread(param[1]) 43 | except: 44 | print ('faild to load %s' % param[1]) 45 | quit() 46 | 47 | if input_img1 is None: 48 | print ('faild to load %s' % param[1]) 49 | quit() 50 | 51 | try: 52 | input_img2 = cv2.imread(param[2]) 53 | except: 54 | print ('faild to load %s' % param[2]) 55 | quit() 56 | 57 | if input_img2 is None: 58 | print ('faild to load %s' % param[1]) 59 | quit() 60 | 61 | gray= cv2.cvtColor(input_img1,cv2.COLOR_BGR2GRAY) 62 | gray2= cv2.cvtColor(input_img2,cv2.COLOR_BGR2GRAY) 63 | 64 | sift = cv2.SIFT() 65 | 66 | kp ,des = sift.detectAndCompute(gray, None) 67 | kp2 ,des2 = sift.detectAndCompute(gray2, None) 68 | 69 | matcher = cv2.DescriptorMatcher_create("FlannBased") 70 | matches = matcher.match(des,des2) 71 | 72 | output_img = drawMatches(gray,kp,gray2,kp2,matches[:100]) 73 | 74 | -------------------------------------------------------------------------------- /tests/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/karaage0703/python-image-processing/0df536fb82883ee160217859f85340d68ff51270/tests/__init__.py -------------------------------------------------------------------------------- /tests/test_extract_color.py: -------------------------------------------------------------------------------- 1 | import pytest 2 | from extract_color import extract_color 3 | import numpy as np 4 | 5 | import cv2 6 | 7 | 8 | # テスト用のダミー画像を生成するヘルパー関数 9 | def create_dummy_image(width, height, color): 10 | image = np.zeros((height, width, 3), dtype=np.uint8) 11 | image[:] = color 12 | return image 13 | 14 | 15 | # テストケース1: h_th_lowがh_th_upより低い場合 16 | def test_extract_color_low_high(): 17 | # 画像を生成（ここでは青色の画像） 18 | src = create_dummy_image(100, 100, (255, 0, 0)) 19 | h_th_low = 50 20 | h_th_up = 100 21 | s_th = 50 22 | v_th = 50 23 | 24 | # extract_colorを実行 25 | result = extract_color(src, h_th_low, h_th_up, s_th, v_th) 26 | 27 | # 想定される結果を計算 28 | expected = np.zeros((100, 100), dtype=np.uint8) 29 | # 青色はHSVでHが約120なので、結果は黒（すべてのピクセルが0）になるはずです 30 | 31 | # 結果のアサーション 32 | assert np.array_equal(result, expected), "The extract_color function did not perform as expected with low < high." 33 | 34 | # テストケース2: h_th_lowがh_th_upより高い場合 35 | def test_extract_color_high_low(): 36 | # 画像を生成（ここでは緑色の画像） 37 | src = create_dummy_image(100, 100, (0, 255, 0)) 38 | h_th_low = 70 39 | h_th_up = 50 40 | s_th = 50 41 | v_th = 50 42 | 43 | # extract_colorを実行 44 | result = extract_color(src, h_th_low, h_th_up, s_th, v_th) 45 | 46 | # 想定される結果を計算 47 | expected = np.zeros((100, 100), dtype=np.uint8) 48 | 49 | # 結果のアサーション 50 | assert np.array_equal(result, expected), "The extract_color function did not perform as expected with high > low." 51 | 52 | 53 | # テストケース3: s_thとv_thの閾値チェック 54 | def test_extract_color_sv_threshold(): 55 | # 彩度が低く、明度が高い色（白に近い色）の画像を生成 56 | src = create_dummy_image(100, 100, (128, 128, 128)) 57 | h_th_low = 20 58 | h_th_up = 30 59 | s_th = 100 # 彩度の閾値を高く設定 60 | v_th = 100 # 明度の閾値を低く設定 61 | 62 | # extract_colorを実行 63 | result = extract_color(src, h_th_low, h_th_up, s_th, v_th) 64 | 65 | # 想定される結果を計算 66 | expected = np.zeros((100, 100), dtype=np.uint8) # 閾値が高いので、全てのピクセルは除外されるべき 67 | 68 | # 結果のアサーション 69 | assert np.array_equal(result, expected), "The s_th and v_th thresholds did not filter out the pixels correctly." 70 | 71 | 72 | # テストケース5: ソース画像がNoneの場合 73 | def test_extract_color_with_none_image(): 74 | with pytest.raises(cv2.error): 75 | extract_color(None, 50, 100, 50, 100) 76 | -------------------------------------------------------------------------------- /watershed.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import cv2 3 | import sys 4 | import numpy as np 5 | # from matplotlib import pyplot as plt 6 | 7 | def watershed(src): 8 | # Change color to gray scale 9 | gray = cv2.cvtColor(src, cv2.COLOR_BGR2GRAY) 10 | 11 | # Use the Otsu's binarization 12 | thresh,bin_img = cv2.threshold(gray,0,255,cv2.THRESH_BINARY_INV+cv2.THRESH_OTSU) 13 | # print(thresh) # print threshold 14 | 15 | # Noise removal 16 | kernel = np.ones((3,3), np.uint8) 17 | opening = cv2.morphologyEx(bin_img,cv2.MORPH_OPEN,kernel,iterations = 2) 18 | 19 | # Sure background area 20 | sure_bg = cv2.dilate(opening,kernel,iterations=3) 21 | 22 | # Finding sure foreground area 23 | dist_transform = cv2.distanceTransform(opening,cv2.DIST_L2,5) 24 | ret, sure_fg = cv2.threshold(dist_transform,0.7*dist_transform.max(),255,0) 25 | 26 | # Finding unknown region 27 | sure_fg = np.uint8(sure_fg) 28 | unknown = cv2.subtract(sure_bg,sure_fg) 29 | 30 | # Marker labelling 31 | ret, markers = cv2.connectedComponents(sure_fg) 32 | 33 | # Add one to all labels so that sure background is not 0, but 1 34 | markers = markers+1 35 | 36 | # Now, mark the region of unknown with zero 37 | markers[unknown==255] = 0 38 | 39 | # Apply watershed 40 | markers = cv2.watershed(src,markers) 41 | src[markers == -1] = [255,0,0] 42 | 43 | # Check marker (If check markers, please import matplotlib) 44 | # plt.imshow(markers) 45 | # plt.show() 46 | 47 | # Check markers data 48 | # print(np.unique(markers,return_counts=True)) 49 | 50 | return markers, src 51 | 52 | if __name__ == '__main__': 53 | param = sys.argv 54 | if (len(param) != 2): 55 | print ("Usage: $ python " + param[0] + " sample.jpg") 56 | quit() 57 | 58 | # open image file 59 | try: 60 | input_img = cv2.imread(param[1]) 61 | except: 62 | print ('faild to load %s' % param[1]) 63 | quit() 64 | 65 | if input_img is None: 66 | print ('faild to load %s' % param[1]) 67 | quit() 68 | 69 | markers, img = watershed(input_img) 70 | cv2.imwrite("watershed_markers_" + param[1], markers) 71 | cv2.imwrite("watershed_image_" + param[1], img) 72 | --------------------------------------------------------------------------------