python 比较两张图片的相似度

版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/Zzhouzhou237/article/details/84033886

import cv2.cv as cv
import cv2
import numpy as np

def compareHist( stdimg, ocimg):
    stdimg = cv2.imread(str(stdimg), 0)
    ocimg = cv2.imread(str(ocimg), 0)
    stdimg = np.float32(stdimg)
    ocimg = np.float32(ocimg)
    stdimg = np.ndarray.flatten(stdimg)
    ocimg = np.ndarray.flatten(ocimg)
    imgocr = np.corrcoef(stdimg, ocimg)
    print imgocr[0, 1]
    return imgocr[0, 1] > 0.96

if __name__ =='__main__':
    stding = 'C:/Users/503061752/Desktop/OCImg.png'
    ocimg ='C:/Users/503061752/Desktop/BJAutoT520/images/Attention.png'
    compareHist(stding,ocimg)

以上代码实现的功能是先读取两张图片，然后比较两张图片的相似程度，当两张图片相似度为0 时，运行到imgocr行时 imgocr 为nan 会报错

numpy\lib\function_base.py:3183: RuntimeWarning: invalid value encountered in true_divide