Infrared Small Target Detection via Non-Convex Rank Approximation Minimization Joint l2, 1 Norm

By Landan Zhang, Lingbing Peng, Tianfang Zhang, Siying Cao, Zhenming Peng. Remote Sensing, 2018

328 citations recorded by the source corpus.

To improve the detection ability of infrared small targets in complex backgrounds, a novel method based on non-convex rank approximation minimization joint l2,1 norm (NRAM) was proposed. Due to the defects of the nuclear norm and l1 norm, the state-of-the-art infrared image-patch (IPI) model usually leaves background residuals in the target image. To fix this problem, a non-convex, tighter rank surrogate and weighted l1 norm are instead utilized, which can suppress the background better while preserving the target efficiently. Considering that many state-of-the-art methods are still unable to fully suppress sparse strong edges, the structured l2,1 norm was introduced to wipe out the strong residuals. Furthermore, with the help of exploiting the structured norm and tighter rank surrogate, the proposed model was more robust when facing various complex or blurry scenes. To solve this non-convex model, an efficient optimization algorithm based on alternating direction method of multipliers (ADMM) plus difference of convex (DC) programming was designed. Extensive experimental results illustrate that the proposed method not only shows superiority in background suppression and target enhancement, but also reduces the computational complexity compared with other baselines.

Subfields: Aerospace Engineering, Electrical and Electronic Engineering, Mechanics of Materials

View Paper: https://www.semanticscholar.org/paper/7d55d7fb4b1d400c7f8c1abaaaed6707d28d1c0a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

1. Zhenming Peng
2. Kobus Barnard
3. Fei Zhou
4. Siying Cao
5. Tianfang Zhang
6. Landan Zhang
7. Yimian Dai
8. Yiquan Wu

In Electrical and Electronic Engineering

1. Zhenming Peng
2. Kobus Barnard
3. Fei Zhou
4. Landan Zhang
5. Yimian Dai
6. Yiquan Wu
7. Wei An
8. Siying Cao

In Mechanics of Materials

1. Zhenming Peng
2. Lei Li
3. Yimian Dai
4. Yiquan Wu

Citations in the corpus

• ISNet: Shape Matters for Infrared Small Target Detection (2022, methodology): “Representative methods include HVS based methods, such as tri-layer local contrast measure (TLLCM) [4] and weighted strengthened local contrast measure (WSLCM) [12], filtering based methods [2, 9], as well as low-rank based methods, such as reweighted infrared patch-tensor (RIPT) [5], partial sum of”
• Dense Nested Attention Network for Infrared Small Target Detection (2021, methodology): “To demonstrate the superiority of our method, we compare our DNA-Net to several state-of-the-art (SOTA) methods, including traditional methods (Top-Hat [6], Max-Median [9], WSLCM [13], TLLCM [12], IPI [7], NRAM [16], RIPT [8], PSTNN [17], MSLSTIPT [5]) and CNN-based methods (MDvsFA-cGAN [32], ACM [2”
• Infrared Small Target Detection Based on Partial Sum of the Tensor Nuclear Norm (2019, methodology): “We can see from the results that the ReWIPI lacks robustness to different scenarios although it does well in Figure 16g. NRAM provides a tighter surrogate of rank with nonconvex rank approximation involved, which implies that the separated background image would be more accurate so that the problem ”
• Attention-Guided Pyramid Context Networks for Detecting Infrared Small Target Under Complex Background (2023, methodology): “…method multiscale patch-based contrast measure (MPCM) [10], matrix-optimization method non-convex rank approximation minimization (NRAM) [16], multisubspace optimized self-regularized weighted sparse (SRWS) [18], and tensor-optimization partial sum of the tensor nuclear norm (PSTNN)…”
• Asymmetric Contextual Modulation for Infrared Small Target Detection (2020, methodology): “Modulation Scheme FPN as Host Network U-Net as Host Network IoU nIoU IoU nIoU b = 1 b = 2 b = 3 b = 4 b = 1 b = 2 b = 3 b = 4 b = 1 b = 2 b = 3 b = 4 b = 1 b = 2 b = 3 b = 4 TopDownLocal 0.595 0.648 0.693 0.713 0.635 0.662 0.688 0.703 0.648 0.710 0.713 0.718 0.673 0.692 0.694 0.697 BiGlobal 0.599 0.”