Infrared small target detection based on multi-directional cumulative measure

Robustness of small target detection is a researchable hotspot in infrared surveillance system. The residual phenomenon background clutter universal current local comparison methods. Algorithm sparse low-rank decomposition restoration cannot be applied to the actual situations due long time consumption. This letter proposes multi-directional cumulative measure (MDCM) enhance saliency and effectiveness weak-small detection. Firstly,multi-directional mean difference implemented central layer estimate background, while derivative multiplying calculated central-active characterize overall target’s heterogeneity, then technology image fusion adopted eliminate interference false target. Finally,a simple adjudicative employed toward separated region from complex scenes. Compared up date existing approaches, extensive simulational testing on four public datasets prove that proposed approach capable separating targets efficiently an irregular single-scale window achieve comparable or even better accuracy.