Solved: Re: OCR to text account key - what is that?

mahakala_ · ‎10-16-2019

Hi I am trying to use flow with powerapps to get OCR to text functionality. While building out in flow, I added the Computer Vision API for the OCR to text, and one of the fields is an account key. I am not sure what that is. I am using a trial version of of powerapps, so I am guessing I need a license for the OCR? Can someone please help. My canvas app will take a picture I have taken with the phone, and it will convert the numbers in the image into text. I am trying to go through flow to do it. If you have other suggestions please advise. Thanks.

mahakala_ · ‎11-01-2019

Hi @PaulD1,

Just wanted to let you know I was able to figure out what caused that error message in flow. So turns in the flow history, it shows my failed flows and there is a generic detailed description explaining why it failed. So error msg was Access denied. You do not have permission to perform this action or access this resource. I thought I didn't have access to sharepoint, but I see in flow that the connection was made to sharepoint. So it turned out that I had the Site address wrong. I was able to find the correct address and it now works. Hoping you can help me answer these two next questions:

1) While i was trying to get this to work, I was trying to get it create file within one drive for business or google drive. So I did get it to work with one drive for business and tried it for at least 5 times, but it started to fail afterwards. I was getting the following msg, wondering if you know what it means (got the same msg for google drive as well):

Unable to process template language expressions in action 'Create_file_2' inputs at line '1' and column '2092': 'The template language expression 'json(decodeBase64(triggerOutputs().headers['X-MS-APIM-Tokens']))['$connections']['shared_onedriveforbusiness']['connectionId']' cannot be evaluated because property 'shared_onedriveforbusiness' doesn't exist, available properties are 'shared_sharepointonline, shared_cognitiveservicescomputervision'. Please see https://aka.ms/logicexpressions for usage details.'.

2) Now since I got the ocr to text to work many thanks to you, sometimes it does not seem accurate. Is there a way to make it more accurate? and is there a way to use regular expressions or filter, to extract just the part of the text that I want?

Many thanks again. I will mark this as solved.

View solution in original post

JoeF-MSFT · ‎10-16-2019

Hi @mahakala_,

AI Builder now offers OCR capabilities in Flow, have a look: https://docs.microsoft.com/en-us/ai-builder/flow-text-recognition

Feel free to ask questions as you use it.

PaulD1 · ‎10-16-2019

Hi @mahakala_

You will need to create an Azure Subscription (don't worry, you can sign up for free and use the free tier of the Computer Vision API) and create a Computer Vision resource (simple, just a few clicks). This will then give you the Account Key that you need to enter into Flow.

I have a step-by-step video showing how to perform OCR from the camera using the Computer Vision API.

If you save an image with text to a modern SharePoint library, there is a 'hidden' column named MediaServiceOcr that will (eventually) populate with the OCRd text - you don't seem to be able to display the column in SharePoint, but you can in PowerApps and MS Access.

AI Builder is another alternative but (as I understand the license model) very expensive and not required for some simple OCR.

mahakala_ · ‎10-17-2019

Hi @JoeF-MSFT ,

Thanks. I will look into this. Someone else told me in this thread that the text account key is for azure which I don't have. Would I need to pay additionally for AI builder if I have powerapps canvas already?

mahakala_ · ‎10-17-2019

Hi @PaulD1,

Thanks for your help. I will sign up for azure and try the free tier of computer vision API and give it a shot. I dont have sharepoint, so can i save the images to say onedrive or cds instead? Do you have vidoes on that, and how do I access your videos?

PaulD1 · ‎10-17-2019

Sorry - forgot to paste in the link: https://www.youtube.com/watch?v=qXSRKfYqYMM&t=3s

It is part of a series of videos that also cover how to set up your Azure subscription (very easy, just a few clicks). The free tier of Computer Vision API gives you plenty of free calls per month to experiment with.

mahakala_ · ‎10-18-2019

Thanks @PaulD1 ,

I will check out your videos. Unfortunately I am having trouble with my trial account now and can't get back into editing the app. I will write you back once I get a chance to play with it.

mahakala_ · ‎10-22-2019

Hi @JoeF-MSFT ,

I finally got my account in working order and tried to follow through the AI Builder as you suggested. But I was not able to get far. I got stuck at step 4. Couldnt find 'Predict - CDS' trigger. Going through AI Builder is this something I have to pay for, or is it included with canvas? Do you have any other documentation for this? I will try to sign up for azure to try their OCR.

JoeF-MSFT · ‎10-22-2019

Hi @mahakala_ ,

The predict action is available in Microsoft Flow, but please note that you will need to create the flow inside a solution: https://docs.microsoft.com/en-us/flow/create-flow-solution

mahakala_ · ‎10-31-2019

Hi @PaulD1,

I watched your video, including the video to create the azure subscription. I went the 7 day route. I set up everything that you had shown in the video, but when I go back to powerapps, it does not see the flow when I click on the button (onselect), action/flows. Only 'create new flow' appears in the data window. In flow, it does see that the flow is there.

Couple things i was not clear on when setting up azure. There are two keys, I only used key 1. Then when setting up the OCR in flow, it asked for the key again, and again I entered key 1. When was I suppose to use Key 2? As for the url, there are 3 and all three are the same https://westcentralus.api.cognitive.microsoft.com/vision.

Hope you can help. Thanks.