Text Detection is poor accuracy.

[HaikunHuang]

I followed the text detection example to make a mobile app that detects the text. However, the accuracy seems poor. As the figure shows, there are many texts on the screen but only a few can be detected. And it seems the number cannot be detected. Any idea to improve the accuracy of the text detection?

[HaikunHuang]

Or, is it possible to train our own model? Thanks.

[EnoxSoftware]

Unfortunately, I do not know how to train this model. https://github.com/opencv/opencv_contrib/commit/c146d37b93a678bc24824de237d0d51b3fa80c7c