Barcode in the page read as gibberish text - need workaround

[Gaurav Shegokar]

I have pdfs where there are barcodes on almost every page, the issue with the current tesseract is that it reads the barcode as gibberish text. 

How do I get around with that? Is it the issue with the line/word segmentation, Or lstm OCR? Or I have to employ some cv tricks to remove barcodes before processing it through tesseract. 

Thank you for your help.

Best,

Gaurav.