在opencv中提供了阈值操作，给定阈值进行分割图像
import cv2 import numpy as np import matplotlib.pyplot as plt img_gray=cv2.imread("c:\\Users\\Administrator\\Desktop\\123\\cat2.jpg",cv2.IMREAD_GRAYSCALE) #>127--->255 <127----->0 ret, thresh1 = cv2.threshold(img_gray, 127, 255, cv2.THRESH_BINARY) #thresh1的反转 ret, thresh2 = cv2.threshold(img_gray, 127, 255, cv2.THRESH_BINARY_INV) #截断值 >127---变为等于255 ret, thresh3 = cv2.threshold(img_gray, 127, 255, cv2.THRESH_TRUNC) #大于127不变，小于等于127变为0---to zero ret, thresh4 = cv2.threshold(img_gray, 127, 255, cv2.THRESH_TOZERO) #thresh4的反转 ret, thresh5 = cv2.threshold(img_gray, 127, 255, cv2.THRESH_TOZERO_INV) titles = ['original', 'binary','binaty_inv','trunc','tozero','tozero_inv'] images = [img_cat2,thresh1,thresh2,thresh3,thresh4,thresh5] for i in range(6): plt.subplot(2,3,i+1),plt.imshow(images[i],'gray') plt.title(titles[i]) plt.xticks([]),plt.yticks([]) plt.show()
上图为读灰度图得到的结果，下图为读原图得到的结果。
上面给了五种方式，上图第一个为原图。改变 cv2.threshold(img_gray, 127, 255, cv2.THRESH_BINARY_INV)的参数可得到不同的结果
1、cv2.threshold(img_gray, 127, 255, cv2.THRESH_BINARY)：像素值大于127的值变为255，小于127的变为0。
2、cv2.threshold(img_gray, 127, 255, cv2.THRESH_BINARY_INV)：与1正好相反。
3、ret, thresh3 = cv2.threshold(img_gray, 127, 255, cv2.THRESH_TRUNC)：像数值大于127变为255，小于127变为不变。
4、ret, thresh4 = cv2.threshold(img_gray, 127, 255, cv2.THRESH_TOZERO)：像数值大于127不变，小于127变为0。
5、ret, thresh5 = cv2.threshold(img_gray, 127, 255, cv2.THRESH_TOZERO_INV)：与4相反，像数值大于127变为0，小于127不变。