pytesseract psm 选项参数

版权声明： https://blog.csdn.net/qq_26877377/article/details/81775000

最近写*车之家的爬虫，遇到动态，扭曲的自定义字符，以前直接比对不变的字符部分已经不行了，想了半天，对字符的操作不是很了解，所以就想到用orc来直接识别好了

遇到问题，使用pytesseract进行操作的时候，添加了中文的语言的选项，但是不添加psm参数时，识别不出来。经过一番查找 找到

应该加上--psm 8 ，将整个图像当初一个汉字来操作

1. Page segmentation modes:

2. 0 Orientation and script detection (OSD) only.

3. 1 Automatic page segmentation with OSD.

4. 2 Automatic page segmentation, but no OSD, or OCR.

5. 3 Fully automatic page segmentation, but no OSD. (Default)

6. 4 Assume a single column of text of variable sizes.

7. 5 Assume a single uniform block of vertically aligned text.

8. 6 Assume a single uniform block of text.

9. 7 Treat the image as a single text line.

10. 8 Treat the image as a single word.

11. 9 Treat the image as a single word in a circle.

12. 10 Treat the image as a single character.

13. 11 Sparse text. Find as much text as possible in no particular order.

14. 12 Sparse text with OSD.

15. 13 Raw line. Treat the image as a single text line,

16. bypassing hacks that are Tesseract-specific.

Here is a sample usage of image_to_string with multiple parameters.

1. target = pytesseract.image_to_string(image, lang='eng', boxes=False, \

2. config='--psm 10 --oem 3 -c tessedit_char_whitelist=0123456789')