In our application, we made use of Computer Vision and OCR to detect text in images.
However, we noticed some text that are parsed are not accurate.
For example, the following text is displayed on the screen: “hostname.root.001”
OCR would return that it detected it as:
“hostname.root.OO1” —> The zero is changed to capital letter ‘O’
“hostname.rootOO1” —> Same changes as above and the last ‘.’ is omitted
UiPath version: 2022.10.07 (Enterprise License)
(1) Are there application-related or project-related settings or options that we can configure
that will improve the detection rate?
(2) Is it possible to train the Computer Vision and OCR? If Yes, we would like to know the procedure how.