Infrared Small and Moving Target Detection on Account of the Minimization of Non-Convex Spatial-Temporal Tensor Low-Rank Approximation under the Complex Background

Infrared point-target detection is one of the key technologies in infrared guidance systems. Due to long observation distance, point target often submerged background clutter and large noise process atmospheric transmission scattering, signal-to-noise ratio low. On other hand, image appears form fuzzy points, so that has no obvious features texture information. Therefore, scholars have proposed many object methods for dimming images, which become a hot research topic on account flow-rank model based patch. However, result high false alarm rate because most low-rank models patch do not consider spatial-temporal characteristics sequences. we introduce 3D total variation (3D-TV) regularize foreground non-convex rank approximation minimization method, as continuity effectively suppress interference caused by dynamic movement extraction. Finally, this paper proposes tensor algorithm (MNSTLA) studying related algorithms detection, experimental results show strong robustness low method compared with advanced algorithms, such NARM, RIPT, WSNMSTIPT.