/\

Package Structure and Key Classes

The application follows a modular package structure organized as follows:

app

• FaceRecognitionApp Face Recognition App
• StudentManager Student Manager System

config

• AppConfig: Manages application configuration settings, including detection parameters and system preferences. All config is always stored and saved into app.properties
• AppLogger: Handles logging throughout the application.
• IConfigChangeListener: Interface for listening to configuration changes.

entity

• Student: Represents student data entities with ID, name, and other attributes.
• Session: Session data, eg(CS102 G1 Week1)
• Roster: Roster data, eg(CS102 G1)

gui (Graphical User Interface)

• config
  • FaceCropSettingsPanel: UI panel for configuring face cropping settings.
• detection
  • Face detection related UI components.
• homepage
  • LoginPage: Main login interface for user authentication.
  • MainDashboard: Primary dashboard after login.
  • SettingsGUI: Settings and configuration interface.
  • ReportsGUI: Reports history interface
• recognition
  • CameraPanel: Displays camera feed and handles video processing.
  • LiveRecognitionViewer: UI for live face recognition functionality.
• settings
  • SettingsCenter: Central hub for managing different settings panels.
  • WelcomeView: Welcome screen in settings.

model

• FaceData: Storing of FaceData of images
• FaceCandidate: A detected Face from a camera
• FaceImage: image and embedding

repository

• DBConnection: Link to Database
• StudentRepositoryInstance: CRUD functionality for Student Database
• SessionRepositoryInstance: CRUD functionality for Session Database
• RosterRepositoryInstance: CRUD functionality for Roster Database
• the rest has relational CRUD functionality for cross-tables

report

• CSVGenerator: Generates CSV reports.
• ExcelGenerator: Generates Excel reports.
• PDFGenerator: Generates PDF reports.
• StudentData: Handles student data for reporting.
• ReportLog: Reporting Logging

service

• embedding
  • FaceEmbeddingGenerator: Generates face embeddings for recognition.
  • EmbeddingQualityAnalyzer: Analyzes quality of generated embeddings.
• recognition
  • LiveRecognitionService: Core service for live face recognition.
• session
  • SessionManager: Manages user sessions.
• student
  • StudentManager: Manages student-related operations.

util

• ImageProcessor: Utility class for image processing operations.
• ColourTheme: Utility class for standardising colours

---

How to Run the Application

Prerequisites

• Java Development Kit (JDK) installed
• OpenCV library files in lib/ directory
• Windows environment (batch files provided)
• Ensure arcface.onnx is downloaded in data/resources (download link: https://huggingface.co/garavv/arcface-onnx)

Steps to Run

1. Compile the Application:

   • Run Compile.bat(Windows) or Compile.sh(mac) to compile all Java source files.
   • This compiles the source from src/ directory to compiled/ directory.

2. Run the Application:

   • Run run.bat(Windows) or run.sh(mac) to start the application.
   • This launches the login page interface. For Admin, username: 123, password: admin123 For TA, username: 456, password: TA123

Notes

• Ensure OpenCV DLL is loaded via System.load(new File("lib/opencv_java480.dll").getAbsolutePath()); in relevant classes.
• VS Code settings should include referenced libraries in .vscode/settings.json.

---

Make openCV .jar and .dll accessible from lib/* for any local machine

1. In .vscode/settings.json, in java.project.referencedLibraries, we call .jar no issue

2. In both all java files, we do:

   System.load(new File("lib/opencv_java480.dll").getAbsolutePath());

   Which is an indirect direct way to get the absolute path of the file.

---

Debugging and Custom Configurations (Developer Settings)

Logs are generated in /logs Custom Configurations and further fine_tuning can be done by modifying app.properties file