A detection method for infrared multi-target in aerospace backgound

Main task of the infrared search and track system is analyzing and identifying targets of airspace. But first this is needed to detect all targets in infrared image. Therefore, the multi-target detection algorithms are studied and we propose an effective multi-target detection method. Firstly, an improved morphological operator is designed based on airspace background and target traits of infrared image. Background is weakened but targets are enhanced when infrared image is processed by the gray morphological filter. Then, potential targets are found by the maximum local sum algorithm. Finally, true targets are affirmed based on data association of sequence images. The infrared images got from long-wavelength infrared camera are processed with the method of the paper. Experiment results show that the method can detect targets in infrared image quickly and accurately.