Single-Frame Infrared Small Target Detection by High Local Variance, Low-Rank and Sparse Decomposition

Infrared Small-Target Detection Based on Background-Suppression Proximal Gradient and GPU Acceleration

“…On one hand, some methods have used prior constraints, including Column-Weighted IPI (WIPI) [18], Non-negative IPI with Partial Sum (NIPPS) [20], and Re-Weighted IPI (ReWIPI) [21]. On the other hand, some studies have identified limitations in the nuclear norm and L1 norm and, so, alternative norms to achieve improved target representation and background suppression have been proposed; for example, Non-convex Rank Approximation Minimization (NRAM) [22] and Non-convex Optimization with Lp norm Constraint (NOLC) [23] introduce non-convex matrix rank approximation coupled with L2,1 norm and Lp norm regularization, while Total Variation Weighted Low-Rank (TVWLR) [24], Kernel Robust Principal Component Analysis (KRPCA) [25] introduce total variation regularization, High Local Variance (HLV) [26] method present LV* norm to constrain the background's local variance. Patch-based methods mainly consider the low-rank nature of the background, affecting their performance in the presence of strong edges.…”

“…Baselines and parameter settings. We compared our proposed method to other stateof-the-art patch-based methods, including IPI [17], NIPPS [20], NRAM [22], NOLC [23], SRWS [34] and HLV [26]. As tensor-based methods have better performance in terms of computational efficiency, we also included three tensor-based methods for comparison, including RIPT [36], PSTNN [38], PFA [37], LogTFNN [39] and ANLPT [42].…”

“…The most popular method is Infrared Patch-Image (IPI) [17], which uses a sliding window technique to generate a corresponding patch image from the original image. Due to its outstanding performance, many studies [18][19][20][21][22][23][24][25][26] have been conducted on IPI, which typically yields superior results. However, patch-based methods still have two problems: (1) The misclassification of strong edges as sparse target components, and (2) the time-consuming nature of the method.…”

Real-Time Recognition Algorithm of Small Target for UAV Infrared Detection

“…At present, small target detection algorithms can be divided into neural network methods [10], space-time tensor methods [11][12][13], and low-rank sparse decomposition [14]. In terms of computational complexity, space-time tensor methods and low-rank sparse decomposition algorithms involve relatively complex mathematical operations and optimization processes.…”