Special Issue on 'Extracting semantics from multi-spectrum video'

摘要

Recent advances in imaging, networking, data processing and storage technology have resulted in an explosion in the use of multi-spectrum image/video in a variety of fields, including video surveillance, urban monitoring, cultural heritage area protection and many others. The integration of videos from multiple channels can provide complementary information and therefore increase the accuracy of the overall decision making process. A typical example is the usage of a hybrid vision system for video surveillance application, which usually consists of omnidirectional, perspective and infrared cameras. Such a system takes advantage of a large view scope from omnidirectional cameras, richer textural information from perspective cameras and better image quality in the dark environment from infrared cameras. However, the benefit of multi-spectrum video fusion always comes with a certain cost and complexity in the analysis process due to the fact that the involved modalities have different characteristics. One of the challenges of such systems is that they need to strike a balance between two opposing requirements. On the one hand, the more pronounced the independence between the different modalities, the more v10.1016/j.patrec.201 1.03.020