Notes
- The projects proposed below have components related to computer vision and pattern recognition. I am not expecting you to have a background in these areas but you should be motivated and enthusiastic about learning new topics.
- Specifically, these projects require strong background in C++ programming and interest in image processing, machine learning, and user interfaces.
- I expect the students to work consistently throughout two semesters and to have periodic meetings with me to discuss their progress.
- Please contact me if you are interested in any of the projects below or would like to propose your own project.
Projects
- Target detection and tracking from multiple sensorsThe goal of this project is to develop a system that captures videos from multiple cameras, enhances these videos, and detects and tracks targets in these videos. The project will be done jointly by Meteksan Savunma and Bilkent University teams. The hardware setup will include a single board computer, a video capture card, cameras with different types, a monitor and other input and output devices. The software will run on embedded Linux and the algorithms will be developed using C++ and Qt libraries in Eclipse. A detailed project description in Turkish is available.
- Image information miningWe are interested in performing content-based classification and retrieval in large archives of satellite images. New generation high-resolution satellite sensors send several terabytes of data to Earth continuously every day. However, current systems allow queries only on metadata such as geographical coordinates, time of acquisition, sensor type, etc. (Have you tried Google Earth?) New techniques and efficient tools are required to perform automatic semantic classification and retrieval based on image content (such as finding satellite scenes with similar land development, analyzing effects of a natural disaster, finding buildings and tracking urbanization patterns, etc.).
- Object recognitionManually annotating objects in large image databases is a very tedious, time consuming and subjective process. New algorithms for automating this process are needed to make the information stored in these databases available and useful. This project involves the design and implementation of new algorithms that use color, shape and context information for large scale object recognition. If successful, the outcomes of both projects can contribute to our existing collaboration with NASA and European Commission researchers.
Students
TBD
Schedule
Please check the list of important deadlines for the 2009-2010 Academic Year.
Related Links
- RETINA Vision and Learning Group
- Information about CS 491: Senior Design Project I
- Information about CS 492: Senior Design Project II
- Computer Vision Home Page
- Pattern Recognition Information Page
.