Users who are not aware of how healthy dishes in a restaurant are can take a photo of its menu. Menual identifies all dishes on the menu, looks dishes up on a food database, and gives feedback and explanations whether each dish is healthy or not according to the German Nutrition Society's (DGE) standards.
Tools used: Android Studio, Firebase, Google Cloud Vision API, Nutritionix API
The Frontend is implemented using Android Studio and Firebase. We use Google Text Detection API to detect the text on a menu. Menual then filters dishes from other text such as ingredients or headlines. It sends all detected menus to the NutritionX API, which returns detailed information on a dish's nutrients and vitamins. Based on these values, Menual calculates a food score for each dish according to the German Nutrition Society's (DGE).
Users can choose from different diet preferences such as low-carb food or vegetarian food and allergies. These preferences influence the score of a dish.
Restaurants that have many healthy dishes and are nearby to the user get recommended.
- Deploy the Firebase code to your Firebase functions
- Get a Nutritionix API key
- Insert your Nutritionix API key in functions/src/index.ts and functions/lib/index.js
- Insert your Google API key in functions/src/index.ts and functions/lib/index.js
- Compile the Frontend code with Android Studio