Road Defect Detection (Smartphone)

Overview

This project introduces an innovative road defect detection system that utilizes smartphone sensors to monitor road conditions in real-time. By analyzing data from the accelerometer and gyroscope, the system can classify road surfaces and generate insights into road quality. This information can be visualized on a map, aiding both drivers and road authorities in decision-making.

Key Features

• Real-Time Road Classification: Detects road irregularities using accelerometer and gyroscope readings from a smartphone.
• Data-Driven Insights: Aggregates sensor data to create a road condition heatmap, identifying areas that need maintenance.
• Machine Learning-Based Classification: Utilizes Support Vector Machine (SVM) and Multi-Layer Perceptron (MLP) models to classify road conditions accurately.
• User-Driven Labeling: Enables manual annotation of road conditions through a mobile app, improving data quality.
• Scalable and Cost-Effective: Operates on smartphones, eliminating the need for expensive hardware installations.

Technical Implementation

• Sensor Data Processing:
  • Collects accelerometer (X, Y, Z) and gyroscope (X, Y, Z) data at 5 readings per second.
  • Uses StandardScaler normalization for improved model performance.
• Machine Learning Models:
  • Support Vector Machine (SVM) for initial classification.
  • Multi-Layer Perceptron (MLP) with three hidden layers for enhanced accuracy.
• Mobile App for Data Collection:
  • Built using Flutter and Dart for a smooth cross-platform experience.
  • Allows users to annotate road conditions, improving dataset reliability.
• Backend & Database:
  • Python backend for data processing.
  • MongoDB for storing road condition data and historical records.

Business Impact

This system provides critical benefits for both individual drivers and government authorities:

• Drivers can receive real-time alerts about road conditions, enabling safer and more comfortable travel.
• Road Authorities can use the aggregated data to prioritize maintenance and reconstruction efforts, leading to cost-effective infrastructure planning.