Skip to content

Local Chapter project

Predicting Road Defects and Optimizing Traffic Light Countdown to Reduce Congestion in Indonesia

Image processingDeep Learning / Computer VisionReinforcement LearningSegmentationObject DetectionEDA
Start date
August 1, 2023
Finish date
September 17, 2023
Project status
completed
Predicting Road Defects and Optimizing Traffic Light Countdown to Reduce Congestion in Indonesia

Challenge background

Traffic management is crucial for preserving road capacity, enhancing safety, and reducing congestion in urban areas like Jakarta, Indonesia. This project aims to leverage advanced technologies and data-driven solutions to address the challenges associated with traffic management in the city.

The problem

Jakarta faces several traffic-related issues, including congestion, accidents, and road defects. These problems lead to inefficiencies, safety hazards, and increased travel times for residents. The primary problems to tackle are: 1. Vehicle speed and category classification to enforce speed limits. 2. Traffic density classification for efficient traffic redirection. 3. Pothole detection for road maintenance and safety.

Goal of the project

  • Develop machine learning models for vehicle category classification and detection.
  • Create a model for traffic density classification.
  • Implement a pothole object detection system.
  • Develop a user-friendly web application for real-time traffic management.
  • Investigate the possibility of future developments like road lane instance segmentation, plate number recognition, and vehicle tracking.

Project timeline

  1. 1

    Week 1

    Project Setup and Data Collection: Set up the project repository and gather relevant datasets.

  2. 2

    Week 2

    Data Preprocessing and Model Planning: Preprocess the data and plan the machine learning models.

  3. 3

    Week 3

    Vehicle Category Classification Model: Develop and train the vehicle category classification model.

  4. 4

    Week 4

    Traffic Density Classification Model: Create the traffic density classification model.

  5. 5

    Week 5

    Pothole Detection Model: Implement the pothole object detection model.

  6. 6

    Week 6

    Web Application Development: Design and develop a user-friendly web application.

  7. 7

    Week 7

    Testing, Documentation, and Future Planning: Conduct testing, document the project, and plan future developments.

What you'll learn

  1. Improved understanding of deep learning model development and evaluation.
  2. Proficiency in using PyTorch, OpenCV, and other relevant libraries.
  3. Experience in creating user-friendly web applications.
  4. Knowledge of traffic management challenges and solutions.
  5. Collaboration and project management skills through teamwork on a complex project.

Get involved

What to expect from a Local Chapter project

First project

  • Welcomes beginners and experienced practitioners.
  • Focuses on education and collaborative delivery.
  • Produces open-source project work.

Benefits

  • Address a significant real-world problem with your skills.
  • Build your project portfolio.
  • Demonstrate your work to organizations and project partners.

Requirements

  • Working English communication.
  • A learning mindset.
  • Commitment to collaborative project work.