python的opencv、pillow、numpy库图片读取、转换、裁剪等常用代码

• 导入opencv、np、pillow模块

import cv2
import numpy as np
from PIL import Image, ImageOps

• pil与opencv库读取、保存、显示图片

# PIL库读取图片
pil_img = Image.open("img.png")
# PIL库保存图片到磁盘
pil_img.save("pil_img.png")
# PIL库显示图片
pil_img.show()

# opencv库读取图片
cv2_img=cv2.imread("img.png")
# opencv库保存图片到磁盘
cv2.imwrite("cv2.png", cv2_img)
# opencv库显示图片
cv2.imshow("cv2_img", cv2_img)
cv2.waitKey(0)
cv2.destroyAllWindows()

• pil与opencv库格式相互转换：

#pil色彩空间转换
pil_img = pil_img.convert('RGBA')

# opencv转pil：cv2是用np数组存储数据的，等同于np转pil
# cv2色彩空间转换,默认是BGR转为RGB
cv2_img = cv2.cvtColor(cv2_img, cv2.COLOR_BGR2RGB)
pil_img = Image.fromarray(cv2_img)

# pil转opencv格式的np数组：
pil_img = Image.open("img.png")
np_img_array = np.asarray(pil_img)
# 色彩空间转换
cv2_img = cv2.cvtColor(np_img_array, cv2.COLOR_RGB2BGR)

• pil与opencv库裁剪图片

# pil裁剪图片：
# 参数分别为：左上角坐标lt_x, lt_y 右下角坐标rb_x, rb_y 
pil_img = pil_img.crop((lt_x, lt_y, rb_x, rb_y))

#opencv即np数组裁剪图片：
np_img_array = np_img_array[lt_y:rb_y, lt_x:rb_x]

• pil与opencv库小图复制到大图中

# pil生成背景图：
pil_bg_img = Image.new("RGBA", (1920, 1080), (255, 255, 255, 0))
# 将小图复制到背景图中
pil_bg_img.paste(pil_img, (x, y))

# opencv即np数组生成背景图：
opencv_bg_img = np.zeros((1920, 1080, 3), dtype='uint8') 
# 替换坐标范围数组数据
opencv_bg_img[lt_y:rb_y, lt_x:rb_x] = cv2_img[lt_y:rb_y, lt_x:rb_x]