eihli / image-table-ocr

Turn images of tables into CSV data. Detect tables from images and run OCR on the cells.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Merging columns are not able to be detected

anhhaibkhn opened this issue · comments

Dear @eihli ,

Thank you very much for your project. It works great!
I have not fully understood your detection algorithms yet, but I think there is this issue, which would be great to improve the accuracy of your package. I noticed that in the case some columns are merged, the program will cut it followed by the major columns. Besides, your program works well in case of rows are merged:
Here is the example:
table_to_cut_vertical

The extract_cell_images_from_table method 's results:

table_type1_indexed10

I will take a look deeper into the code, meanwhile, I think it's better to report this to you so that the library can be enhanced in the future.
Asides from this minor issue, your library is awesome.

Thanks again and best regards

tb1
tb5
It's not so good. It failed with those 2 imgs