python opencv 检测图像全白

Question

How to detect if an image is white. It means that the image has only a white background. This is for black background. Example code:

image = cv2.imread("image.jpg", 0)
if cv2.countNonZero(image) == 0:
    print "Image is black"
    return True
else:
    print "Colored image"
    return False

Answer 1

You can perform a bitwise_not operation on the input image and apply the same logic(this is just a hack):

image = cv2.imread("image.jpg", 0)
image = cv2.bitwise_not(image)

if cv2.countNonZero(image) == 0:
    print "Image is white"
    return True
else:
    print "Black region is there"
    return False

Answer 2

You could just usenumpy.all :

Full white image:

img_white = np.ones([10, 10, 3], np.uint8) * 255

res1 = np.all([img_white == 255])
print(res1) #=> True

Non full white image:

img_non_white = img_white.copy()
img_non_white[1, 1] = (100, 255, 255)

res2 = np.all([img_non_white == 255])
print(res2) #=> False

Answer 3

If you read the image with OpenCV, then it comes as a numpy array. So what you need to do is to check how many white points are in there.

import cv2
import numpy as np

img = cv2.imread('img.png', cv2.IMREAD_GRAYSCALE)
n._white_pix = np.sum(img == 255)