NutroMate

NutroMate classifies the brands of the packaged product instantly. To begin with, it classifies only brands of the Bread and Fruit. NutroMate uses transfer learning concept which allows model creation with significantly reduced training data and time by modifying existing rich deep learning models. NutroMate's data model is retrained MobileNet for product brand images. MobileNets are a class of convolutional neural network designed by researchers at Google. They are coined “mobile-first” in that they’re architected from the ground up to be resource-friendly and run quickly, right on the phone.

As the user points towards the brand of the product, percentage of prediction and product name is shown from the retrained model. It starts showing the the prediction results from 80% match and awaits till the prediction is 90% to show the product selection action sheet. Once the product is selected from the actionsheet user will be taken to Rating Page which shows the Health Star Rating of the product calculated for serving of 100g. "NutritionDetails" call to action will take user to nutrition details of the product. Nutrition details show the average nutritional content of Energy, Protein, Fat, Carbohydrate and Sodium in the product for avergae serving size.

HealthStar rating is a quick and easy way to compare the nutritional profile of packaged foods. It helps you to make healthier choices.

Nutritional Information gives clearly readable nutrient related information for the average servicng size.

We have tried to cover most of the popular bread brands, but it is possible that your favorite brand is not in the NutroMate's database. In that case you can send us the image of the bread by clicking "Report" in the app.