Aerial Infrared Dim Target Detection Method

The weak and small targets detection in infrared images is one of the key techniques in infrared imaging guidance systems, and it is also widely used in public application, such as the astronomy prognosticates, particle collision, forest warning and remote sensing etc. As a result, the weak and small targets detection in IR images has strong value in both theory and application.Since background in the IR images changes quite slowly, it is meaningful to detect targets by background prediction. But the performance of background prediction is influenced with the background fluctuation.In this paper, background prediction based targets detection methods are studied. Then two modified methods are proposed to detect aerial targets. One is based on a new bright and dim point classification around edge area, and another is a multi-scale edge neighborhood prediction. Theory analysis and computer simulation indicate that the two methods can work better in background inhibition and targets enhancement, and eliminate false alarm points caused by the edge of cloud more effectively.The detection of moving weak and small targets in IR sequences is discussed and studied finally, based on a neighborhood judging method. Emulation results show the method requires lower computation and can distinguish weak and small targets from sequences effectively and reliably.