UiPath Screen OCR: Now in Public Preview!

UiPath Screen OCR: Now in Public Preview!

UiPath Screen OCR is our own in-house, machine-learning based OCR targeted for screens and digital text.

It can be used as an alternative to the other OCR engines, with any of the available Screen Scraping activities from the UI Automation or Computer Vision packages. And the final scope is to become the default Computer Vision text detection engine.

From our tests, Screen OCR performs much better on application screens & digital text than all leading OCR engines

Bold claim, right? We challenge you to prove us wrong! :wink: Do a quick test with an existing workflow (or build something new, either works) and drop our Screen OCR engine in, then compare the difference.

Please share your experiences below. We’re excited to see everyone put this to the test!

How to Get Started

  1. Open your Manage Packages window
  2. Click “Official” in the left navigation
  3. Click the “Include Prerelease” checkbox
  4. Search for “UiPath.ScreenOCR.Activities” and install the latest version

Feel free to also review the activity guide for reference.

60-second Demo

55 Likes

i am getting the error Thread aborted while installing

@Cosin

After installing it is asking for API key.

I am not able to use the feature.

Please help me in this regard.

Get the Computer Vision api from http://platform.uipath.com/ .
Add the api in that activity while your using computer vision scope activity no need to add api other wise you getting error add computer vision api

2 Likes

Could you give more details about what you were doing to get this error?

1 Like

If you want to use just the OCR, you don’t need an API key.

If the API key is for ComputerVison, yes you need to create an account and get the key from cloud.uipath.com (or platform), from the Licenses tab, and add it to the CV Scope.

1 Like

Dont know its a bug or something … For the First time i got error like thread aborted and asked me to restart … After that it worked

@Cosin I tried it by leaving it as blank. When it was not working the I pasted the API key from platform.uipath.com which is also not working in my case.

@Vijay_RPA Can you please help me in this regard?

@Syed_Kashif_Nasar Please send screenshort or XML file.

Dear @pradeepRPA

Please find the attachment.

Capture2

Send the CV screen scope and uipath screen OCR properties and another thing you didn’t indicate the screen in cv screen scope activity.please check that.

2 Likes

You probably didn’t paste the API key in the CV Scope, or you pasted something else.
You need to get the key from Platform, licenses page, Other services, ComputerVison

Good Morning,

I added this to an existing workflow that utilizes the read pdf activity. I switched the regular read pdf activity for the read pdf with OCR and utilized this new OCR engine. It more or less didn’t read the data and provided ‘003/15PAGE’
as the output.

Hope this information will help with future iterations/updates.

Thanks,

image007.png

image008.png

image009.png

image010.png

image011.jpg

image012.jpg

Please share the properties of CV Screen scope It is missing API key from Cloud.

regards,

It’s amazing,I think it’s the best OCR Engine,Because I was puzzling in CAPTCHA,this Engine can recognition some CAPTCHA after little adjustment!
Thank you Uipath!

2 Likes

Could you pls post a screenshot of the workflow, and at least (but preferably a whole page) screenshot of the PDF? Is it a scanned document or digital?

I cannot find Official tab in our orgnization’s Studio (installed on my machine). Any idea how to get it added ? Official tab missing and UiPath.ScreenOCR.Activities cannot be seen

1 Like

Wow, that’s a beast OCR. Just what I needed long ago. Thank you Uipath!

2 Likes

Getting null ocr text.
I have added following key and url.
API ky: “MiwKrgC6/dQeUE64NijtHuxmzGL63+/ZEEXknaM1PUu78bbZGTAuIs+SrH9RkzCo”
url: “https://cv.uipath.com/

@Cosin

1 Like