How does Tesseract OCR fuse results from multiple models? #81
Description
In the latest Tesseract, the -l option allows for, say, eng+chi_sim. How does Tesseract OCR fuse results from multiple models? Can someone point to the code that does the fusion? Thanks!