Question

我是Python新手。我试图识别来自TensorFlow的图像中的文本。当我尝试在终端中运行python gen.py时，我收到了TypeError。

Traceback (most recent call last):
  File "gen.py", line 287, in 
   for img_idx, (im, c, p) in enumerate(im_gen):
  File "gen.py", line 277, in generate_ims
   yield generate_im(random.choice(char_ims), num_bg_images)
  File "gen.py", line 239, in generate_im
   bg = generate_bg(num_bg_images)
  File "gen.py", line 226, in generate_bg
   bg = cv2.imread(fname, cv2.IMREAD_GRAYSCALE) / 255.
TypeError: unsupported operand type(s) for /: 'NoneType' and 'float'

Answer 1

看起来您的代码无法在该路径上打开图片

bg = cv2.imread(fname, cv2.IMREAD_GRAYSCALE) / 255

cv2.imread方法返回None。

最好先拍摄图像

image = cv2.imread(fname, cv2.IMREAD_GRAYSCALE)
if image is not None:
    bg = image / 255

您也可以在开头导入os模块以检查该文件是否存在

import os
# <your code...>
if !os.path.isfile(fname):
    raise Exception("File doesn't exist")

image = cv2.imread(fname, cv2.IMREAD_GRAYSCALE)
if image is not None:
    bg = image / 255

TypeError：/：＆＃39; NoneType＆＃39;不支持的操作数类型并且＆＃39;漂浮＆＃39;

1 个答案: