Splitting the image in blocs, paragraphs, lines and chars.

[Vinetos]

For each image, after being processed, images must be split in blocs containing the text.

Inside blocs, we must detect:

• [x] paragraphs
• [x] lines
• [x] words
• [x] characters

These little matrix will be then sent to the neural network.

[starcruiser5289]

need more research for paragraphs segmentation. but code works on basic document such as word(it just does not recognise paragraphs) all code will be updated to return a text once linked to neural network