John Smith

So, in version 4.6.3 Capture2Text we have -- leptonica 1.74.4 and tesseract 4.00 , which represented in "pvt.cppan.demo.danbloomberg.leptonica-1.74.4.dll" and "tesseract400.dll" Will it work, if we will replace those dll on a build, made by those instructions? -- https://bucket401.blogspot.com/2021/03/building-tesserocr-on-ms-windows-64bit.html

Here is actually a way to fix this. https://sourceforge.net/p/capture2text/tickets/218/

Here is actually a way to fix this. https://sourceforge.net/p/capture2text/tickets/218/

It would be so great.

Simplified Chinese -- Poor and inaccurate

Here's "debug_capture" and "debug_enhanced" results. It's okay, "gImageReader-3.4.0" works with images of this type pretty bad too. https://github.com/manisandro/gImageReader/releases/tag/v3.4.0

Here's "debug_capture" and "debug_enhanced" results. It's okay, "gImageReader-3.4.0" works with шmages of this type pretty bad too. https://github.com/manisandro/gImageReader/releases/tag/v3.4.0

And I'm not talking about the module with hundreds of languages in "Easy Screen OCR" that uses Google Vision API, I'm talking about built-in module with ten languages, that doesn't use Google Vision API -- it performs recognition of such text without a single error. I think it's just that all open-source enthusiasts at the moment believe that "Capture2Text" works also well so nobody fixed it. But it doesn't work well and it's sad.