Skip to content

Welcome to the Risky Building Detection using Deep Learning on Satellite Images GitHub repository! This project aims to address the alarming issue of risky buildings in Dhaka city, Bangladesh, by leveraging deep learning technology to detect and identify these buildings from satellite images. By accurately identifying risky buildings,

Notifications You must be signed in to change notification settings

sftSalman/Risky_building_detection_of_dhaka_city_from_satelite_images-

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

27 Commits
 
 
 
 
 
 

Repository files navigation

Code and paper will be available soon

Dhaka Risky Buildings Detection with YOLOv5

Overview

Bangladesh, among the most densely populated countries globally, faces a critical issue with the structural safety of buildings in its capital city, Dhaka. Shockingly, approximately 90% of the buildings in Dhaka are deemed highly risky and susceptible to collapse during natural disasters like earthquakes. To address this pressing concern, our team aims to develop a solution utilizing deep learning technology to identify risky buildings within Dhaka city using satellite imagery.

Project Objective

The primary goal of this project is to leverage the power of the YOLOv5 object detection model to analyze satellite images of Dhaka. By employing deep learning techniques, we aim to pinpoint and classify buildings that are at high risk of collapse during natural hazards. The ultimate objective is to provide valuable insights to local authorities, enabling them to identify high-risk areas promptly and take necessary measures to ensure public safety.

Methodology

  • Utilization of YOLOv5 model for object detection and classification
  • Collection and preprocessing of satellite images of Dhaka city
  • Training the model to recognize and categorize risky buildings
  • Validation and testing of the model's accuracy and reliability
  • Generating actionable insights and risk assessment reports based on the model's predictions

How to Contribute

We welcome contributions from the community to enhance the accuracy and effectiveness of our model. Whether through refining the training dataset, improving the model architecture, or suggesting innovative approaches, your contributions are valuable in our quest to make Dhaka city safer.

Future Scope

  • Integration of real-time satellite data for continuous monitoring
  • Collaboration with local authorities for effective implementation
  • Expansion to other cities or regions facing similar structural safety challenges
  • Enhancing model robustness for varied environmental conditions

Get Involved

Feel free to reach out, open issues, or submit pull requests to support this initiative. Together, we can leverage technology to mitigate the risks posed by structurally vulnerable buildings in Dhaka, ensuring a safer environment for all residents.

1 Fig : Satelite view of Dhaka city

About

Welcome to the Risky Building Detection using Deep Learning on Satellite Images GitHub repository! This project aims to address the alarming issue of risky buildings in Dhaka city, Bangladesh, by leveraging deep learning technology to detect and identify these buildings from satellite images. By accurately identifying risky buildings,

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published