0
\$\begingroup\$

i am trying to develop image recognition iOS app , i will have one reference image , and user has to draw something one screen and when user release finger , then i will convert his drawing into UIImage/CGImage, then i will be comparing the percentage of closeness to reference image .

If the percentage will be more then 90% then user will win . (See picture below , ignore 🔴 Red(user Canvas ) and 🔶 Yellow border(reference boundary will not include in atul image ) ) enter image description here

What i have done till now :

I create custom image array of each reference and then create MLModel from this website

https://cloud.ibm.com/developer/appledevelopment/starter-kits/custom-vision-model-for-core-ml-with-watson

I tried to do this with the MLModel but every-time , app is showing 90+% percentage of , i am not able to do this .

Please suggest

I am trying to use MLKit in iOS Swift , i am doing wrong , if yes then what is the correct way to do this .

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.