Dear Encodian Flowr Team,
I am writing to suggest an enhancement to the "Apply OCR (AI)" action. Currently, the OCR process extracts noise such as random characters from natural backgrounds (e.g., trees and ground) in images. I propose the implementation of a feature that allows the OCR to discern and exclude these elements while still extracting meaningful text from things like words on t-shirts, street signs and car plates.
This feature would significantly improve the clarity and usefulness of the text extracted from such images.
Thank you for considering this improvement.
Best Regards,
Reno Sun
