Infrared Small Target Detection Utilizing the Enhanced Closest-Mean Background Estimation

Background estimation is an efficient infrared (IR) small target detection method. However, to deal with unknown targets, the window in existing algorithms should be adjusted perform multiscale and requires a lot of calculations. Besides, stages during after have received wide attention algorithms, but research on before insufficient. Moreover, typically regard maximum value different orientations as value. when dim adjacent high-brightness background, it easily submerged. This article proposes three-layer detect targets sizes only single-scale calculation. The enhanced closest-mean background method then proposed carefully designed before, during, estimation. Before estimation, matched filter adopted improve image signal-to-noise ratio. During principle suppress background. After ratio-difference operation performed enhance true simultaneously. A simple checking mechanism further performance. Experiments some IR images demonstrate effectiveness robustness Compared has better enhancement, suppression, computational efficiency.