Tables recognition using Google Cloud Vision API

[Дмитрий Ковнер]

Hello! I use OCR function (DOCUMENT_TEXT_DETECTION) of Google Vision API to process different medical documents. There are some tables in them. According to Google's documentation there is a special BlockType for tables (https://cloud.google.com/vision/docs/reference/rpc/google.cloud.vision.v1#blocktype), however, I can not get this sort of block in a response from Google even in a case when existence of a table is obvious. Does anybody know the reason of such behavior? Do I need to use special options or methods to detect tables using Google Vision API? An example of a table I tried to process is in the attachment.

[Carlos Recoder Moreno]

Hello Parikshit,

Block represents one "logical" element of the page—for example, an area covered by text, or a picture or separator between columns. The text and table blocks contain the main information needed to extract the text. Currently the ability to detect tables in images via Cloud Vision API OCR is not available. 

However, a Public Issue Tracker Report [1] for that feature has already been created that allows you to monitor its progress by simply starring it. 

Kindly note that there is no ETA for when this feature will be releases. All future updates regarding this matter will be communicated in the public report.

___

[1]: https://issuetracker.google.com/117948935