How does Textract deal with actual text in a PDF?

0

How does Textract deal with PDF documents that contain actual text, as in text that you can copy and paste into a Notepad document (as opposed to an image that Textract can recognise text in). Does Textract simply take the text verbatim, or does it render the text as an image, and then OCR it?

Topics

Machine Learning & AI

Tags

Amazon Textract

Language

English

asked 3 years ago413 views

1 Answer

Newest
Most votes
Most comments

1

Accepted Answer

Textract does the latter where it first renders the PDF as an image and then performs OCR on it.

AWS-User-7817432

answered 3 years ago

EXPERT

reviewed a year ago

Relevant content

Bedrock versus Textract for document text/meaning extraction
Accepted Answer
Scott
asked a year ago
Training Textract to Separate Text Blocks Into Separate Components with a Delimiter
SCHIFAWS
asked 6 months ago
Textract does not recognize image text
rePost-User-1218412
asked 2 years ago
Extracting data from PDF that contains strikeout text using Amazon Textract in Python
SomebodySysop
asked 4 months ago
How do I make sure that the email that I received is actually from Amazon?
AWS OFFICIALUpdated 2 years ago
Is the text message that I received from AWS?
AWS OFFICIALUpdated 2 years ago
Why can't I copy and paste in WorkSpaces?
AWS OFFICIALUpdated a month ago
How can I record my Linux terminal session in text mode and share or replay it later?
AWS OFFICIALUpdated a year ago
How do I pause a queue and cancel thousands of jobs in that queue?
EXPERT
Bo_L
published a year ago
How to use the AWS CLI to collect Cost Explorer Forecasting data per account for all accounts in an AWS Organization
EXPERT
Fernando_F
published a year ago