Visual Odometry with Drone Images

Visual Odometry with Drone Images

Suggested Team Size: 2-3 students

Duration: 2 semesters


Project Description
The Visual Odometry with Drone Images project is a cutting-edge research and development initiative that seeks to advance the field of autonomous drone navigation. Leveraging computer vision and image processing techniques, this project aims to develop a visual odometry system that enables drones to navigate, map their surroundings, and estimate their position and orientation in real-time using onboard cameras and imagery.

Project Objectives
The primary objectives of the Visual Odometry with Drone Images project are as follows:

1. Visual Odometry Algorithm Development: Research, design, and implement state-of-the-art visual odometry algorithms tailored to the unique challenges and constraints of drone imagery.

2. Quick Image Processing: Develop efficient image processing pipelines capable of handling large volumes of high-resolution images in real-time, ensuring responsive drone navigation.

3. Autonomous Navigation: Enable drones to autonomously navigate in various environments, including urban, rural, and indoor settings, using only visual information from onboard cameras.

4. Robustness to Environmental Variability: Ensure the visual odometry system can handle changes in lighting conditions, weather, and environmental factors, maintaining reliable performance.

5. Sensor Fusion: Integrate data from other sensors, such as GPS, with visual odometry to enhance navigation accuracy, particularly in GPS-denied environments.

6. Machine Learning Enhancements: Explore the use of machine learning techniques, such as deep neural networks, for feature extraction, image registration, and error correction.

7. User Interface: Develop a user-friendly interface that allows operators to monitor and control the drone's navigation and mapping capabilities.

8. Testing and Validation: Conduct extensive field tests and simulations to validate the accuracy, reliability, and robustness of the visual odometry system across various scenarios and terrains.

9. Documentation and Knowledge Transfer: Provide comprehensive documentation and training resources to enable others to understand, deploy, and extend the visual odometry system.

The Visual Odometry with Drone Images project represents a significant advancement in autonomous drone technology. By developing a reliable and accurate visual odometry system, this project will open up new possibilities for applications in agriculture, construction, search and rescue, environmental monitoring, and beyond, where drones can navigate and map with precision, even in challenging and GPS-denied environments.




Project Advisor: 

H. Birkan Yılmaz

Project Status: 

Project Year: 

  • Spring

Contact us

Department of Computer Engineering, Boğaziçi University,
34342 Bebek, Istanbul, Turkey

  • Phone: +90 212 359 45 23/24
  • Fax: +90 212 2872461

Connect with us

We're on Social Networks. Follow us & get in touch.