├── ImgSegment.py
└── README.md


/ImgSegment.py:
--------------------------------------------------------------------------------
  1 | #!/usr/bin/env python2
  2 | # -*- coding: utf-8 -*-
  3 | """
  4 | Created on Wed Jul 24 10:39:21 2019
  5 | 
  6 | @author: nageshsinghchauhan
  7 | """
  8 | #SImple image segmentation using K-Means clustering algo
  9 | 
 10 | #color clustering
 11 | #Image segmentation from video using OpenCV and K-means clustering
 12 | import numpy as np
 13 | import cv2
 14 | import matplotlib.pyplot as plt
 15 | original_image = cv2.imread("/Users/nageshsinghchauhan/Desktop/images/p.jpg")
 16 | # Converting from BGR Colours Space to HSV
 17 | img=cv2.cvtColor(original_image,cv2.COLOR_BGR2RGB)
 18 | vectorized = img.reshape((-1,3))
 19 | # convert to np.float32
 20 | vectorized = np.float32(vectorized)
 21 | # Here we are applying k-means clustering so that the pixels around a colour are consistent and gave same BGR/HSV values
 22 | # define criteria, number of clusters(K) and apply kmeans()
 23 | criteria = (cv2.TERM_CRITERIA_EPS + cv2.TERM_CRITERIA_MAX_ITER, 10, 1.0)
 24 | # We are going to cluster with k = 2, because the image will have just two colours ,a white background and the colour of the patch
 25 | K = 4
 26 | attempts=10
 27 | ret,label,center=cv2.kmeans(vectorized,K,None,criteria,attempts,cv2.KMEANS_PP_CENTERS)
 28 | # Now convert back into uint8
 29 | #now we have to access the labels to regenerate the clustered image
 30 | center = np.uint8(center)
 31 | res = center[label.flatten()]
 32 | res2 = res.reshape((img.shape))
 33 | #res2 is the result of the frame which has undergone k-means clustering
 34 | figure_size = 15
 35 | plt.figure(figsize=(figure_size,figure_size))
 36 | plt.subplot(1,2,1),plt.imshow(img)
 37 | plt.title('Original'), plt.xticks([]), plt.yticks([])
 38 | plt.subplot(1,2,2),plt.imshow(res2)
 39 | plt.title('K = %i' % K), plt.xticks([]), plt.yticks([])
 40 | plt.show()
 41 | 
 42 | #canny edge detection
 43 | edges = cv2.Canny(img,100,200)
 44 | plt.figure(figsize=(figure_size,figure_size))
 45 | plt.subplot(1,2,1),plt.imshow(img)
 46 | plt.title('Original Image'), plt.xticks([]), plt.yticks([])
 47 | plt.subplot(1,2,2),plt.imshow(edges,cmap = 'gray')
 48 | plt.title('Edge Image'), plt.xticks([]), plt.yticks([])
 49 | 
 50 | plt.show()
 51 | 
 52 | """
 53 | #video
 54 | import cv2
 55 | import numpy as np
 56 | 
 57 | # Create a VideoCapture object and read from input file
 58 | cap = cv2.VideoCapture("/Users/nageshsinghchauhan/Documents/bb.mkv")
 59 | 
 60 | # Check if camera opened successfully
 61 | if (cap.isOpened()== False):
 62 |   print("Error opening video  file")
 63 | 
 64 | # Read until video is completed
 65 | while(cap.isOpened()):
 66 |   # Capture frame-by-frame
 67 |   ret, frame = cap.read()
 68 |   if ret == True:
 69 |     # Display the resulting frame
 70 |     cv2.imshow('Frame', frame)
 71 |     # Press Q on keyboard to  exit
 72 |     if cv2.waitKey(25) & 0xFF == ord('q'):
 73 |       break
 74 |   # Break the loop
 75 |   else:
 76 |     break
 77 | # When everything done, release
 78 | # the video capture object
 79 | cap.release()
 80 | # Closes all the frames
 81 | cv2.destroyAllWindows()
 82 | """
 83 | 
 84 | 
 85 | 
 86 | 
 87 | #plotting an Image in 3D color space.
 88 | import matplotlib.pyplot as plt
 89 | from mpl_toolkits.mplot3d import Axes3D
 90 | import cv2
 91 | 
 92 | #read image
 93 | img = cv2.imread("/Users/nageshsinghchauhan/Documents/bb.mkv")
 94 | 
 95 | #convert from BGR to RGB
 96 | img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
 97 | 
 98 | #get rgb values from image to 1D array
 99 | r, g, b = cv2.split(img)
100 | r = r.flatten()
101 | g = g.flatten()
102 | b = b.flatten()
103 | 
104 | #plotting
105 | fig = plt.figure()
106 | ax = Axes3D(fig)
107 | ax.scatter(r, g, b)
108 | plt.show()
109 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
 1 | # Image-segmentation-K-means-clustering
 2 | 
 3 | Image segmentation is an important step in image processing, and it seems everywhere if we want to analyze what’s inside the image. For example, if we seek to find if there is a chair or person inside an indoor image, we may need image segmentation to separate objects and analyze each object individually to check what it is. Image segmentation usually serves as the pre-processing before pattern recognition, feature extraction, and compression of the image.
 4 | 
 5 | 
 6 | Image segmentation is the process of partitioning a digital image into multiple distinct regions containing each pixel(sets of pixels, also known as superpixels) with similar attributes.
 7 | 
 8 | For more information, checkout this article.
 9 | 
10 | 
11 | https://www.theaidream.com/post/introduction-to-image-segmentation-with-k-means-clustering
12 | 


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