I have been using Tesseract to OCR images and it has been working very well. I would like to use this same library to convert tables, however I havent found a way to do it online.
My thought was to locate the horizontal and vertical lines in a bitmap, section those off, save each as a new image, and then OCR the images before compiling them back into a string to produce the table.
My question(s):
1) Is this method valid, or is there a better (easier) free way/library I can use to complete this?
2) How exactly could I go about searching the bitmap for vertical/horizontal/intersections to create my pixel range for the cell?
Any help is greatly appreciated!
~Frab
My thought was to locate the horizontal and vertical lines in a bitmap, section those off, save each as a new image, and then OCR the images before compiling them back into a string to produce the table.
My question(s):
1) Is this method valid, or is there a better (easier) free way/library I can use to complete this?
2) How exactly could I go about searching the bitmap for vertical/horizontal/intersections to create my pixel range for the cell?
Any help is greatly appreciated!
~Frab