food image detection

Question

Hi, I am a college student working on a mobile app project, and I am new to Azure.

I am looking for an appropriate API to detect food and label it from a photo having several food items.

What I have found is :

1. Object detection from Custom vision API - but I had to provide many labeled images (15 images per label).
2. Detect API from cognitive service - I wonder how detail label it would give me.

I don't need exact actual food labels, but it can be general categories like 'fruit' for banana and strawberry, 'vegetable' for any vegetables.

Also, I would not be able to train the model with a lot of images since it is one semester long project.
I wonder if there is anyway I can use Custom vision API without training guessing Azure already has a trained model for general objects.

Or I wonder how useful the detect API is and if it works well with various food.

Please help me!
Thanks

Answer 1

Hi, thanks for reaching out. Analyze Image API would be useful for your scenario. No need label images, the Computer Vision service provides you with AI algorithms for processing images and returning information on their visual features. You can call this API through native SDK or REST API calls. Include Objects in the visualFeatures query parameter, you will get the full JSON response with with the detected objects, and you can parse the string for the contents of the objects section. Let me know if this helps or if you need further guidance, thanks!