Architectures and Algorithms for Multi-source Object Detection and Tracking

Target tracking using multiple information sources can generally provide better performance and reliability than using a single one. The paper proposes a new solution for multi-source information collection and retrieval. Two processing architectures and sensors for information collection are presented: cameras to image sequences, and radio transmitters to RFID information. Algorithms related to the information fusion between sensors are discussed. Test results show that the proposed solution is capable to track objects with a good degree of precision. Abrupt changes of lighting conditions can also be handled, especially when photometric invariant color features are used.