Question

4 Please use python and opencv library in python to implement the following tasks. Given a gray image 'test.jpg', first use Average filter to smooth it, then detect keypoints by using Harris corner detector, finally show the image with keypoints. (10.09)

Answer 1

Below is the python3 code for mentioned problem statement:

import numpy as np
import cv2 as cv
from matplotlib import pyplot as plt

if __name__ == '__main__':
    # read image using opencv
    img = cv.imread('Port_noisy_big.png')
    grayImage = cv.cvtColor(img, cv.COLOR_BGR2GRAY)
    # find out kernel by avaraging
    kernel = np.ones((5, 5), np.float32) / 25
    # Apply avarage filtering
    noiseFreeImage = cv.filter2D(grayImage, -1, kernel)

    # Applying Harris Corner algo
    grayImage = np.float32(grayImage)
    cornerHarrisImage = cv.cornerHarris(grayImage, 2, 3, 0.04)
    # Dilate the for making the corners
    harrisCorner = cv.dilate(cornerHarrisImage, None)
    # applying threshold for an optimal value.
    img[harrisCorner > 0.01 * harrisCorner.max()] = [0, 0, 255]
    cv.imshow('Harris Corrner', img)

    # plot the images
    plt.subplot(121)
    plt.imshow(grayImage), plt.title('Original')
    plt.xticks([]), plt.yticks([])
    plt.subplot(122)
    plt.imshow(noiseFreeImage)
    plt.title('Averaging')
    plt.xticks([])
    plt.yticks([])
    plt.show()

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 import numpy as np import cv2 as cv from matplotlib import pyplot as plt if name _main_': # read image using opencv img = cv.imread('Port_noisy_big.png') grayImage = cv.cvtColor(img, cv.COLOR_BGR2GRAY) # find out kernel by avaraging kernel = np.ones((5, 5), np.float32) / 25 # Apply avarage filtering noiseFreeImage = cv.filter2D (grayImage, -1, kernel) # Applying Harris Corner algo grayImage = np.float32(grayImage) cornerHarris Image = cv.cornerHarris(grayImage, 2, 3, 0.04) # Dilate the for making the corners harrisCorner = cv.dilate(cornerHarris Image, None) # applying threshold for an optimal value. img[harrisCorner > 0.01 * harris Corner.max()] = [0, 0, 255] cv.imshow 'Harris Corrner', img) # plot the images plt.subplot(121) plt.imshow(grayImage), plt.title('Original') plt.xticks(11), plt.yticks([]) plt. subplot (122) plt.imshow(noiseFree Image) plt.title('Averaging') plt.xticks([]) plt.yticks(1) plt.show() 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

Figure 1 Harris Corrner > #QWA Original Averaging (X=7, y=75) - R:113 G:114 B:118