Babatunde

Anomaly Detection App

A robust object detection system designed to identify anomalies among students, using Covenant University as a case study with some base detection classes.

Anomaly Detection App

🧠 Project Overview

The primary objective of this project is to develop a robust object detection system designed to identify anomalies among students, using Covenant University as a case study with some base detection classes.

Object detection is a critical technology that underpins a wide range of modern applications, including:

  • Surveillance systems
  • Security monitoring
  • Autonomous vehicles

Its role in enhancing safety, security, and operational efficiency is indispensable. This project demonstrates the power of AI in real-world scenarios where anomaly detection can prevent incidents and improve overall campus safety.


⚙️ Tech Stack

  • Frontend: Next.js
  • Programming Languages: Python, TypeScript
  • Machine Learning Frameworks: TensorFlow, PyTorch
  • Object Detection Models: YOLO, Faster R-CNN
  • Data Handling: OpenCV, NumPy, Pandas
  • Deployment: Vercel
  • Visualization: Matplotlib, Seaborn, and custom dashboard

🏗 Features & Capabilities

  • Detects unusual or unexpected student behaviors in video feeds
  • Supports multiple base detection classes for common objects/events
  • Real-time alerts for anomalies
  • Data visualization for tracking and analysis
  • Scalable architecture for large datasets

🚀 Challenges & Learnings

  • Handling imbalanced datasets for anomaly detection
  • Reducing false positives while maintaining high detection accuracy
  • Optimizing inference speed for real-time video streams
  • Integrating computer vision models into a functional monitoring system

## 📸 Gallery

image

image


🔗 Links


🏁 Summary

The Anomaly Detection App leverages advanced object detection algorithms to enhance security and operational safety. By applying AI to real-world campus scenarios, it provides actionable insights and real-time monitoring, illustrating the transformative impact of machine learning in anomaly detection.