Google Vision API - Can we have custom preprocessing logic included with detect document text and use that directly as a Vision API

Question

I am using Google Vision API to get all the text scraped out of my Image. However I want to extract text only in a specific area on the image and not the entire text by resizing the image. However for all images the area is not defined as a fixed co-ordinate; the images are error screenshots where the message could be in any side and sometimes the screenshots are taken by having multiple windows open on the monitor. I want to use custom logic to first process the image to identify the section where error message could be present and segment only that portion and then pass that portion of image for text extraction. However I dont want to have any preprocessing logic outside of Vision API. Is there a way where we can include additional logic for preprocessing within Vision API framework and get a api that can then perform both preprocessing and text extraction ?

Answer 1

The answer is quite short: No

The google vision test-detection API is quite short and simple and does not provide any way to do what you requested.