找出哪个图像的黑色最多

Question

首先，我将视频帧分为三个不同的帧。然后我将该图像转换为灰度图像。转换后，我需要找到哪一帧比其他两帧的黑色最多。我不知道如何找到图像中存在多少黑色。我已经在互联网上使用了一些方法，但我不知道如何实现它们？import cv2import syssys.path.append('/usr/local/lib/python2.7/site-packages')import cv2import numpy as npimport timeimport copycap = cv2.VideoCapture(0)while True:    ret, frame = cap.read()    output = frame.copy()    gray = cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)    cv2.imshow('gray',gray)    cv2.imshow('frame',frame)    gray = cv2.cvtColor(output, cv2.COLOR_BGR2GRAY)    gray =255-gray    ret, thresh = cv2.threshold(gray, 225, 255, cv2.THRESH_BINARY_INV)    output, contours, hierarchy = cv2.findContours(thresh, cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)    output = cv2.drawContours(output, contours, -1,(0,0,255),3)    cv2.imshow('frame',output)    height, width = output.shape[:2]    print (output.shape)    start_row, start_col = int(0), int(0)    # Let's get the ending pixel coordinates (bottom right of cropped top)    end_row, end_col = int(height), int(width*.3)    cropped_top = output[start_row:end_row , start_col:end_col]    print (start_row, end_row)     print (start_col, end_col)    cv2.imshow("Cropped Topp", cropped_top)     # Let's get the starting pixel coordiantes (top left of cropped bottom)    start_row, start_col = int(0), int(width*.3)    # Let's get the ending pixel coordinates (bottom right of cropped bottom)    end_row, end_col = int(height), int(width)    cropped_bot = output[start_row:end_row , start_col:end_col]    print (start_row, end_row )    print (start_col, end_col)    ##cv2.imshow("Cropped Bot", cropped_bot)     ##cv2.waitKey(0)     ##cv2.destroyAllWindows()    start_row, start_col = int(0), int(0)    # Let's get the ending pixel coordinates (bottom right of cropped top)    end_row, end_col = int(height), int(width*.3)    cropped_top = cropped_bot[start_row:end_row , start_col:end_col]    print (start_row, end_row)     print (start_col, end_col)    cv2.imshow("Cropped Top", cropped_top) 我需要知道如何检测图像中的黑色并将它们与其他两个图像相关联，并找出哪一个具有最多的黑色

Answer 1

如果像素的相应值为零，则认为黑白图像中的 Hi 为黑色。因此，假设您使用 open cv 和 numpy 数组进行处理，我们可以做这样的事情。

import cv2

import numpy as np

img = cv2.imread('pathOfImg',0) #read img as b/w as an numpy array

unique, counts = np.unique(img, return_counts=True)

mapColorCounts = dict(zip(unique, counts))

现在mapColorCounts[0]将是图像中的黑色像素数，这个数字越多，图像越黑。