How marcoxmarquez/scaling-couscous works
Project Type and Purpose
• This is a native Android mobile application (“VerificaMTC”) written in Java/Kotlin DSL for Gradle.
• It serves as a client-side identity verification tool for driving exams in Peru, integrating:
• RENIEC identity validation (external government API)
• On-device facial recognition using TensorFlow Lite (local biometric authentication)
• Administrative and user dashboards
• Real-time data synchronization via Firebase
File Structure Analysis
• Top-level “app” module: Android Studio project with Gradle build scripts (build.gradle.kts, settings.gradle.kts).
• src/main/java/com/master/verificamtc: subdivided into packages—
• auth: UI activities for login/registration flows
• admin: admin dashboard and face recognition activities
• user: user dashboards, payment, schedule, vehicle management
• helpers: data-access layer (Firebase helpers, local SQLite helper), face-recognition engine wrappers
• utils: cryptographic utilities (BCrypt, AES encryption)
• src/main/assets: TFLite models (facenet.tflite, mobile_face_net.tflite)
• Firebase configuration in google-services.json and related helpers (FirebaseDatabaseHelper, FireDatabaseStore)
• Local persistence via LocalDatabaseHelper (SQLite or Room)
Key Components
a. Mobile App (Android Device)
• UI Layer: Activities/Fragments in auth, admin, user packages
• Business Logic Layer: ViewModels/controllers coordinating flows
• Data Layer:
• Local Database (SQLite) – LocalDatabaseHelper
• On-device Face Recognition – TFLiteFaceRecognition, FaceClassifier, FaceDatabaseHelper
• Remote Backend – Firebase Auth, Firebase Realtime Database / Firestore (via FirebaseDatabaseHelper)
b. External Services
• RENIEC API (RESTful Government identity validation)
• Firebase Cloud Services (Auth, Database, Storage, Cloud Functions if any)
c. Security Components
• Encryption Module – SecurityHelper (AES/BCrypt for sensitive data at rest and in transit)
Relationships & Data Flows
• Startup/UI → Business Logic → Data Layer
• Authentication Flow: App ←→ Firebase Auth
• Registration Flow:
• App → RENIEC API (identity check)
• App → Firebase Database (store user profile)
• Biometric Flow:
• Camera input → FaceDetectionActivity → TFLiteFaceRecognition → local face embeddings store (FaceDatabaseHelper)
• Decision → App → Firebase (update authentication status)
• Real-time Sync: Data changes in Firebase Database ↔ App UI via listeners in FirebaseDatabaseHelper
• Local Persistence: Caching user/session data in SQLite to support offline scenarios
Architectural Patterns & Principles
• MVC/MVVM flavor: separation of UI (Activities) and data/business logic (helpers, classifiers)
• Repository pattern in helpers – abstracts data source (local vs. remote)
• Modular package structure: auth, admin, user, helpers, utils
• Secure data handling: encryption utilities, secure storage of credentials