Unanswered

ComputerVisionAPI V2 Method OCRV2 and OCRTextV2 missing parameters

(0) Share

Report

Posted on by matremblay

I'm trying to use the ComputerVision API with the OCRV2 or OCRTextV2 as the original OCR and OCRText methods are no longer available (unless I'm missing something).

The issue I'm having is that the API documentation for OCRTextV2 gives somes parameters:

https://docs.microsoft.com/en-us/connectors/cognitiveservicescomputervision/

Required Parameters
Image Source
string
Source of the image - either included or by reference url.
Key:
format
Image
dynamic
Key:
Image

But the PowerApps web studio gives me different ones and there is no way to insert the image in the request:
See First image

But it seems to work fine with DescribeContentV2
See second image.

Any idea what is wrong ?

Categories:

Building Power Apps

All responses (3)

Answers (0)

Sort by

v-xida-msft Microsoft Employee on at

Like
a
(1)

Report
Copy link

Link copied!

Hi @matremblay ,

Do you want to use the ComputerVisionAPI.OCRV2() function and the ComputerVisionAPI.OCRTextV2() function within your canvas app?

Based on the issue that you mentioned, I have made a test, and the issue is confirmed on my side. Currently, within PowerApps, when you use ComputerVisionAPI.OCRV2() function or the ComputerVisionAPI.OCRTextV2() function, there is no place to provide the Image argument (second argument).

I think this issue may be related to the Computer Vision API (the Image parameter is not be exposed within OCRV2 action and OCRTextV2 action). Currently, within PowerApps, I afraid that there is no way to achieve your needs.

If you would like this feature to be improved within PowerApps, please submit an idea to PowerApps Ideas Forum:

https://powerusers.microsoft.com/t5/PowerApps-Ideas/idb-p/PowerAppsIdeas

As an alternative solution, you could consider achieve your needs using Microsoft Flow. I have made a test on my side, please take a try with the following workaround:

In addition, you could also consider take a try to create a custom connector based on Computer Vision API within your app to achieve your needs. Please check and see if the following blog or video would help in your scenario:

https://powerapps.microsoft.com/en-us/blog/cognitive-services-with-powerapps-using-custom-connectors/

https://www.youtube.com/watch?v=KKon0Yqwu38

If you capture the picture within your canvas app using Camera control instead of Add Picture control, I think the following video would also achieve your needs:

https://www.youtube.com/watch?v=qXSRKfYqYMM

Best regards,
matremblay 3 on at

Like
a
(0)

Report
Copy link

Link copied!

I see, IO will post a suggestion to try to push them to update the API.
One thing, I'm not sure how to send the captured image from my Powerapp to the Microsoft Flow then trigger it, then retrieve the informations. Can you point me to the right direction ?

hongkong

277 on at

Like (0)

Report

Copy link

Link copied!

OCRTextV3 have been released for preview. Is it possible to use OCRTextV3 in Power App? Can we pass "Image" parameter and what is "Dynamic" type means in Power App?

Name Key Required Type Description

Resource Subdomain or Region	subdomainName	True	string	Azure subdomain or geographic region (e.g. westus)
Image Source	format	True	string	Source of the image - either included or by reference url.
Image	Image	True	dynamic