AI-based Recipe Generator from Ingredient Photos

About the project

Deciding what to cook with the few vegetables and groceries at home is a daily problem, and recipe websites need a dish name first. This AI recipe generator Android project (RasoiLens) starts from a photo instead: Google Gemini Vision lists the ingredients, the user corrects them with simple chips, and Gemini writes recipes for the chosen food type, cooking time and cuisine. The app itself checks every AI recipe, removes unsafe ones (for example meat for a Veg user) and shows the match % and missing items, so the user knows what can be cooked right now.

How it works

• Take or pick a photo of the ingredients; the app checks the file and resizes it
• Gemini Vision detects the items and the app maps them to an ingredient master that understands names like kanda, batata and kothimbir
• Choose Veg / Eggetarian / Non-veg / Jain, cooking time and cuisine
• Gemini writes recipes in JSON; the app checks food type and time and calculates match % and missing items
• No AI key or internet? The app matches your ingredients with approved community recipes instead
• Open a recipe to see nutrition, steps and tips, add it to the weekly meal plan or mark it cooked with a rating
• Make the shopping list from missing items or from the whole week plan

Users

• Home Cook: scans ingredients, gets AI recipes, saves favourites, plans meals, keeps a shopping list
• Chef: publishes recipes with photos after admin verification and sees approval status, remarks and reviews
• Admin: verifies chefs, approves or rejects recipes, manages the ingredient master and Gemini settings, prints reports

Highlights

• 3 roles, 18 Android screens, 14 admin web pages, 28 PHP API files, 12 MySQL tables
• Gemini Vision ingredient detection plus AI recipe generation with app-side safety checks
• Offline community matching so the app works without AI
• Weekly meal plan, smart shopping list with voice typing, recipe and meal plan PDF
• Every form validated in the app and again in PHP (email, 10-digit mobile, password, dates, photos, numbers)
• Secure: hashed passwords, prepared statements, token-based APIs, CSRF protection in admin forms
• Sample data and demo logins for home cook, chef and admin

What you get

• Complete Android Studio project (Java) and PHP admin panel source code
• MySQL database (.sql) with sample data
• Project report (Word, 60+ pages) with UML, DFD and ER diagrams
• Project abstract (2 pages) and project explanation with viva questions
• Setup guide (PDF) and installation help

Who can use this project

Suitable as a final year project for BE / B.Tech (Computer, IT), MCA, BCA and Diploma students who want an Android project with source code that uses generative AI. It is a good choice for an AI and machine learning based mobile app project with documentation and report.

Frequently asked questions

• Is a Gemini API key required? A free key from Google AI Studio enables photo detection and AI recipes; without it the app still gives matching community recipes.
• Can I run it on a real phone? Yes. Keep the phone and computer on the same Wi-Fi and enter the computer IP on the login screen.
• Does it handle Jain and vegetarian food? Yes. The prompt includes the rules and the app also removes recipes that break them.
• Which database and server are used? MySQL with PHP on XAMPP; the database file is included.