Computer Vision(CV)

Hi,
I am in need to open a pdf and get the text. However, I can’t use Anchor base activity or any other.
Can anyone please help me with CV?
Thanks…

Hi @abivanth.r

Use Read pdf activity then by using Regex you can get the required value

Hi @abivanth.r

=> Use Read PDF Text or Read PDF with OCR if the PDF is scanned and store the output in a variable.
=> You can use regular expressions to extract the particular text you want.

Regards

The pdf is imaged and the language in that is chinese or japnese. i can’t use OCR too… data is getting deducted

@abivanth.r

Try with Tesseract OCR it supports that languages

image

Cheers!!

used it too… i can get the data for one pdf. but having multiple pdf’s with structured data. i can’t use OCR… it’s not getting correct data.

OCR never will. It’s not going to be 100% accurate.

Computer Vision is not a solution for this. Computer Vision is about identifying screen elements like buttons, text boxes, etc.

Document Understanding is what you want to use, but again it’ll still be OCR and not 100% accurate.