By using AWS re:Post, you agree to the AWS re:Post Terms of Use
AWS re:Postre:Post

Textract doesn't detect few lines from top of the page.

0

Hi,

Sometimes Textract does not extract OCR from the top of the page. The following shows an example: Enter image description here The input image is 300 DPI. Is this a bug or I am missing a pre-process step or setting?

Thanks

Topics
Machine Learning & AI
Tags
Amazon Textract
Language
English
AWS-User-5556436
asked 2 years ago235 views
1 Answer
0

Thanks for bringing up the issue. 300 DIP seems like a reasonable resolution at which OCR should work fine. However, I can confirm some of the details at the top are missing. So, in this case, could you please reach out to the Textract team via a support case citing quality issue - and provide the redacted document to help the team debug what exactly is happening with the document. Thanks.

AWS
Rohan_K
answered 2 years ago
  • AWS-User-5556436
    a year ago

    Hi,

    I found there is a bug in Textract API. If the page has a barcode or QR code a the bottom of the page, It won't pick up few lines from the top. If I remove the barcode from the page, then it will report back all the the text in the document.

    Thanks.

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions

Relevant content