Replies: 3 comments 7 replies
-
Beta Was this translation helpful? Give feedback.
-
|
谢谢。怎么转换成RapidUnWrap这样的项目，方便调用？
|
Beta Was this translation helpful? Give feedback.
-
这个目前暂时只能你自己想办法了，就是写代码，目前我这边有其他事情很忙，未来可能会集成到RapidUnwrapped里面 |
Beta Was this translation helpful? Give feedback.
-
|
完全不会写。只能等你更新RapidUnwrapped |
Beta Was this translation helpful? Give feedback.
-
|
检测可视化时可以看到有文本行，但是检测+识别可视化时，有的文本行消失了，这是哪个参数控制的？确保所有检测到的文本行在识别可视化也有结果？ |
Beta Was this translation helpful? Give feedback.
-
|
这个没办法保证完全可行，使用多语言识别模型试试呢，没有参数能保证都能识别哈 |
Beta Was this translation helpful? Give feedback.
-
|
好像是方向识别错误导致的，单独分2步，遍历好像可以解决 |
Beta Was this translation helpful? Give feedback.
-
|
改成这个可以解决text_score=0.01, |
Beta Was this translation helpful? Give feedback.
-
|
ok |
Beta Was this translation helpful? Give feedback.
-
|
单字符识别得分多少，才能确保正确？==1？ |
Beta Was this translation helpful? Give feedback.
-
除了RapidUnWrap，有没有其他前处理模型或代码，提高单行识别准确率？将原图处理成无noise的高清晰白底黑字图？
Beta Was this translation helpful? Give feedback.
All reactions