1、下载label.exe文件
链接：github地址
2、安装一下anaconda，百度一下直接安装就行
3、打开labelme.exe文件，直接加载图片，然后编辑多边形，就是mask的位置

4、画好mask了，保存为json文件，记住这个文件夹路径
5、将json文件转换为mask图像，命令行运行：python test.py

替换一下里面的路径即可，然后就生成了mask图像了

import json
import numpy as np
import cv2

# read json file
with open("./501.json", "r") as f:
    data = f.read()

# convert str to json objs
data = json.loads(data)

# read image to get shape
image = cv2.imread("./501.jpg")

# create a blank image
mask = np.zeros_like(image, dtype=np.uint8)

# loop through all shapes in the JSON data
for shape in data["shapes"]:
    # get the points for each shape
    points = shape["points"]
    points = np.array(points, dtype=np.int32)  # tips: points location must be int32
    
    # fill the contour with 255 for each shape
    cv2.fillPoly(mask, [points], (255, 255, 255))

# save the mask
cv2.imwrite("./mask.png", mask)