Find a better binarization method for color images. Produce HTML output. Deal with broken characters. Make a better layout detector. Separate (more) merged characters. Deal better with frames, lines, pictures, noisy characters, etc.
Find a better binarization method for color images. Produce HTML output. Deal with broken characters. Make a better layout detector. Separate (more) merged characters. Deal better with frames, lines, pictures, noisy characters, etc.