test.py

import numpy as np
import scipy.integrate as sci_int
import math
import cv2
from matplotlib import pyplot as plt

# def plot_feature_points(img, feturePoints):
#     # Show image
#     plt.figure(1)
#     plt.imshow(img, cmap='gray', interpolation = 'bicubic')
#     plt.title('Gray-scale image'), plt.xticks([]), plt.yticks([])
#     # Plot feature points
#     pX, pY = feturePoints
#     # Note: Axes of plot and image is reverted
#     plt.plot(pY, pX, '*')

# Read img
grayImg = cv2.imread(r"C:\Users\Tien Hao\Google Drive\Junior\TGMT\TH\DataSet\TestImages\02.jpg", cv2.IMREAD_GRAYSCALE)

# # Convert type as np.float
# grayf = np.float32(grayImg)
# # Call cv2.cornerHarris function
# dst = cv2.cornerHarris(grayf,2,3,0.04)

# # Threshold image
# x, y = np.nonzero(dst > 0.05*dst.max())

# plot_feature_points(grayImg, (x, y))

# plt.show()

# for i, b in enumerate(bin_list):
#     print(i)
#     c[b] += flata[i]

# c[bin_list] += flata
# print(c)

# # Declare sift() obj
# sift = cv2.xfeatures2d.SIFT_create()
# # Sift detector
# keypoints = sift.detect(grayImg, None)

# # Flags will have draw cirle with scale of keypoint and its orientation
# img = cv2.drawKeypoints(grayImg, keypoints, flags=cv2.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS)

# cv2.imwrite('sift_keypoints.jpg',img)

print("The number of matching keypoints between 2 images are %d." %len(grayImg))