How to fetch x,y coordinates of specific pixels(blue colored) from an image in python?-CodePudding

My task is to get the x,y coordinates from the segmented image(blue dots) as produced by the code. How do I automate this process? My end results should be a zip of x,y coordinates of these blue dots as produced in the second image.

Code to generate the blue points:

    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
# import cv2_imshow
# from google.colab.patches import cv2_imshow
 
    image = cv2.imread('./S3/frame35.jpg')
#cv2_imshow(image)

    result = image.copy()
 
    image = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)

    # lower boundary RED color range values; Hue (0 - 10)
    lower1 = np.array([0, 100, 20])
    upper1 = np.array([10, 255, 255])

    # upper boundary RED color range values; Hue (160 - 180)
    lower2 = np.array([160,100,20])
    upper2 = np.array([179,255,255])

    lower_mask = cv2.inRange(image, lower1, upper1)
    upper_mask = cv2.inRange(image, lower2, upper2)

    full_mask = lower_mask   upper_mask;

    result = cv2.bitwise_and(result, result, mask=full_mask)

    plt.figure(figsize=[20,20])
    plt.axis("off")
#     plt.subplot(121);plt.imshow(image[:,:,::-1]);plt.title("Original Image",fontdict={'fontsize': 25});plt.axis('off');
    plt.subplot(122);plt.imshow(result, cmap='gray');plt.title("Mask of red Color",fontdict={'fontsize': 25});plt.axis('on');
    plt.savefig('mask_1.jpg', bbox_inches = 'tight')
#     cv2_imshow(full_mask)
#     cv2_imshow(result)
    #print(full_mask)
    #print(result) 
    cv2.waitKey(0)
    cv2.destroyAllWindows()

CodePudding user response：

This is rather straightforward to do by computing the connected components of a binary image.

First you'd have to threshold your image to get a binary image that segments the dots. You can do this for instance with cv2.threshold. Then you can use cv2.connectedComponentsWithStats, this will - among other things - return a list of the centroids for all the components. One of them will be the background though, but among the returned values there is an integer arrray of the same size where all components have a different label. So you can look up the label of the component of the background and remove that from the centroid coordinates.

EDIT: The code you added does not even run, but I feed it your second screenshot as a mask.png and use the mentioned commands as follows, and the plot the centroids over the image:

src = cv2.imread('mask.png')
_, thresh = cv2.threshold(src[:, :, 0:1], 120, 255, cv2.THRESH_BINARY)
_, _, _, centroids = cv2.connectedComponentsWithStats(thresh, 8, cv2.CV_32S)