Process only text contained in a certain bounding box

0

Hello, in our application we want to process only the name and address info that is handwritten into a form.
It seems that we can parse the output to eliminate all but what we want by using the bounding box info.

But... would it be possible to:

  1. focus the tool only on the bounding box info
  2. tell the tool what is in each box on the form, i.e. name, street address, city, state, zip
  3. have the tool learn and get better at returning valid results by applying rules to each box.

Thanks

JimboCA
已提問 3 年前檢視次數 211 次
2 個答案
0

Unfortunately, there is currently no way to do that. You can try to crop the document to focus on a smaller section of the document, but Textract works best on full document images. I have added a feature request to support (2) in the future

AWS
已回答 3 年前
0

THANK YOU

JimboCA
已回答 3 年前

您尚未登入。 登入 去張貼答案。

一個好的回答可以清楚地回答問題並提供建設性的意見回饋,同時有助於提問者的專業成長。

回答問題指南