Skip to content

Latest commit

 

History

History
157 lines (99 loc) · 6.64 KB

README.md

File metadata and controls

157 lines (99 loc) · 6.64 KB

7ff81f98-da83-4d4c-a8b2-d9e480476e30

PotSafe

Project Description

PotSafe is a pothole detection system that uses image processing techniques to automatically detect locate determine the severity of potholes on roads. The goal of this project is to develop an automated system that can detect potholes in real-time and alert authorities or maintenance crews to repair them, thereby improving road safety and reducing the cost of road maintenance.

PowerPoint Presentation

Abstract for PotSafe

Trained Model Weights

Demo Video

2-D Detection Demo

sections.mp4

PotSafe is a pothole detection system that utilizes image processing techniques to automatically detect, locate, and determine the severity of potholes on roads. The system can be integrated into vehicles equipped with cameras, and can detect potholes in real-time as the vehicle travels.

The main goal of the PotSafe project is to develop an automated system that can detect potholes in real-time and alert authorities or maintenance crews to repair them. By detecting potholes early and alerting maintenance crews in a timely manner, the system aims to improve road safety and reduce the cost of road maintenance.

Overall, PotSafe is a valuable tool for improving road safety and reducing the cost of road maintenance. It is a complete solution for pothole detection, location and severity detection and repairs tracking. It can be integrated with other systems like a map-based app for citizens to report potholes, and a management portal for the road maintenance team to track the progress and repair history.

Depth Detection

Input WhatsApp Image 2023-01-21 at 11 20 14 PM WhatsApp Image 2023-01-21 at 11 19 39 PM

Output

WhatsApp Image 2023-01-21 at 11 20 14 PM (1)

WhatsApp Image 2023-01-21 at 11 19 38 PM

Acknowledgments

Scope of Improvement -

  • Increasing Accuracy of Model by training on better dataset and more types of models like Faster-RCNNs.
  • More accuracy and calculation feature along with MiDAS model for 3D-reconstruction .




PotSafe is under MIT License, Please Read the LICENSE


Technologies Used -

OpenCV Python TensorFlow Keras Git

Authors -

Aryan Raj

Aryan Raj

ML and CV Developer

Sayyad Abid

Aryan Raj

IoT and ML Enthusiast

Arnav Sharma

Arnav

AI/ML Practitioner

Sarthak Vashisht

Sarthak Vashisht

ML and CV Developer