Is there a way to set specific regions on an image for OCR?

I have screenshots of data that I want to export to Excel using OCR. My data is in row/column format, and I would like to keep the formatting after it goes through OCR. Is there a way that I can set specific regions like rectangles on the screenshot that would correspond to cells on Excel?

I know 'roi' allows you to set a rectangle on the image to be processed. I am wondering if it would be possible to have multiple roi's on an image. Each roi would be a cell on Excel.

Answer 1

[Anh Tran]

You can specify region of interest, ROI, as the second argument to OCR function. You can set the region of interest to cover cell, column, or entire matrix on the pictures. This design decision depends on how your data look like. I would recommend you to try different approaches and determine which one is best for your data.