方法一：
# import os
# import cv2
# img_input_path = 'image/1.jpg'
# filename = img_input_path.split("/")[-1].split(".")[0]
# image = cv2.imread(img_input_path)
# region = image[int(520):int(1580), int(338):int(713)] # [338, 520, 713, 1580]
# cv2.imwrite('image/' + filename + '_0.jpg', region)
方法二：
from PIL import Image
x0 = max(int(x1-(x2-x1)*0.05),0)
x00= min(int(x2+(x2-x1)*0.05),w)
y0= max(int(y1-(y2-y1)*0.05),0)
y0= min(int(y1+(y2-y1)*0.05),h)
img = Image.open(img_path)
img = img.convert('RGB')
region = img.crop((x0,y0,x00,y00))
region.save(save_path+filemane+"_0.jpg", quality=95)
版权声明：本文为qq_42262610原创文章，遵循CC 4.0 BY-SA版权协议，转载请附上原文出处链接和本声明。