import cv2 # [1]导入OpenCv开源库import numpy as np image_path = "F://11111111111111111111111111111//100000.jpg"srcImg = cv2.imread(image_path) # [2]将图片加载到内存 cv2.namedWindow("[srcImg]", cv2.WINDOW_AUTOSIZE) # [3]创建显示窗口cv2.imshow("[srcImg]", srcImg) # [4]在刚才创建的显示窗口中显示刚在加载的图片cv2.waitKey(0) # ========================================================================================================# 模块说明:# 由于OpenCv中,imread()函数读进来的图片,其本质上就是一个三维的数组,这个NumPy中的三维数组是一致的,所以设置图片的 # ROI区域的问题,就转换成数组的切片问题,在Python中,数组就是一个列表序列,所以使用列表的切片就可以完成ROI区域的设置 # ========================================================================================================image_save_path_head = "F://11111111111111111111111111111//111//cat_ROI_"image_save_path_tail = ".jpg"seq = 1for i in range(2): # [1]480*360==15*11---height for j in range(2): # [2]column-----------width img_roi = srcImg[(i * 112):((i + 1) * 112), (j * 112):((j + 1) * 112)] image_save_path = "%s%d%s" % (image_save_path_head, seq, image_save_path_tail)##将整数和字符串连接在一起 cv2.imwrite(image_save_path, img_roi) seq = seq + 1 
