Image recognition to text ocr issue

Questions about Thunkable Web API's
, ,
1

hello, i’m a thunkable beginner. really looking forward for help right now. So the application i’m making is basically taking a picture using camera then the system will use image recognition to check text in the picture and make it an output for the label that will be put in the menu.

im having a big issue with OCR since the one i tried isn’t working even after checking few times. everything seem to be fine. im really looking forward for help or advices on this as i’m a beginner and isn’t that much used to this website.

helps will be appreciated if possible, this work is really rush since ive been working for it for the past weeks and i have to finish it by monday. thanks you so much. pictures below containing whats in the blocks and component so far.

image

image
image859×414 42.3 KB

2

sorry if it confusing but to make it short. im struggling to make an application which is image recognition to text by using a camera. im mainly having issue with the blocks and OCR but the rest is kinda fine.

3

There are two things we need to be able to help you: 1. a description of exactly what is/is not working and 2. a screenshot of your blocks or a link to your project. You’ve only provided #2.

I also recommend with any API that you assign the green response block to a text input’s text and then copy and paste the response and post it here. Make sure to format that text using the </> button in the forum toolbar. Once we have that, we can tell you if you’ve parsed the JSON correctly.

4

So basically, The output on the text suppose to be the text that on the image which in this example, It’s suppose to show “Hello” but after i tested it. The output show “null”

I tried to do as you said which put the response block in the text’ input’s text instead. The second picture is what it shows

image
image

5

What I am suggesting is that you preview the project and then copy the text from the text input (be sure to select all) and then post it here as text (not a screenshot).

6

it says
{"ParsedResults":[{"TextOrientation":"0","FileParseExitCode":1,"ParsedText":"","ErrorMessage":"","ErrorDetails":""}],"OCRExitCode":1,"IsErroredOnProcessing":false,"ProcessingTimeInMilliseconds":"1296","SearchablePDFURL":"Searchable PDF not generated as it was not requested."}

7

It looks like you got a valid response and the text it found was none.

image
image1136×682 55.6 KB

So you might have to try a different image or a different OCR API. Many free OCR APIs are very poor quality and don’t recognize text very well.

Try this url instead of the Image1’s Picture and let me know the JSON response: https://www.mockofun.com/wp-content/uploads/2020/04/stop-sign-5614.jpg

image
image850×637 60.6 KB

Because using their test GET call, it works. This url: https://api.ocr.space/parse/imageurl?apikey=helloworld&url=https://www.mockofun.com/wp-content/uploads/2020/04/stop-sign-5614.jpg returns this JSON:

image
image1166×974 85.8 KB

8

I tried and it worked, Thanks you so much!

9

Did it work with your image or with the sample image url I gave you?

10

Both.