Prior-Guided Data Augmentation for Infrared Small Target Detection

Wang, Ao; Li, Wei; Huang, Zhanchao; Wu, X.; Jie, Feiran; Tao, Ran

doi:10.1109/jstars.2022.3222758

Cited by 13 publications

(5 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(1) Datasets: We conducted experiments on IRSTD-1k [18] , SIRST dataset [20] and MSISTD [42] respectively. SIRST contains 427 infrared images.…”

Section: A Experimental Settingsmentioning

confidence: 99%

See 1 more Smart Citation

Generative Adversarial Differential Analysis for Infrared Small Target Detection

Ma,

Pang,

Hao

2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Infrared small target detection refers to the extraction of small targets in a complex, low signal-to-noise ratio background. Depthwise convolution makes it difficult to comprehensively characterize small infrared targets and ignores the importance of image background for the detection task. In this paper, we propose the generative adversarial differential analysis (GADA) model for infrared small target detection, the core of which aims to weaken the reliance on target features and enhance the use of background information. Specifically, we first construct pseudo background labels by the fast marching method. Then, the background-guided generative adversarial network is used to learn the background data distribution. On this basis, the differential image containing interest regions of small targets is obtained by differential analysis. Finally, the detection results are obtained by performing an elaborate characterization of the interest regions. The effectiveness of GADA is verified with three public datasets. Compared to several state-of-the-art methods, GADA achieves better performance in terms of F 1, IoU , and AU C.

show abstract

“…(1) Datasets: We conducted experiments on IRSTD-1k [18] , SIRST dataset [20] and MSISTD [42] respectively. SIRST contains 427 infrared images.…”

Section: A Experimental Settingsmentioning

confidence: 99%

“…MSISTD contains 1077 images. It greatly expands the context of singleframe small target detection from being limited to the detection of aerial targets only to the detection of multiscene, multiscale, and lower SCR targets [42]. For three datasets, they are divided into training and testing sets in the ratio of 8:2.…”

Section: A Experimental Settingsmentioning

confidence: 99%

Generative Adversarial Differential Analysis for Infrared Small Target Detection

Ma,

Pang,

Hao

2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…To capture more target features, Sun et al [35] propose receptive-field and direction-induced attention network (RDIAN). To alleviate data scarcity, some researchers attempt to augment data [36], [37] or fuse traditional feature representations [38]- [40]. For the space-based infrared tiny ship detection, Wu et al [41] propose multilevel TransUNet (MTU-Net) that achieves better multilevel feature extraction.…”

Section: A Infrared Small Target Detectionmentioning

confidence: 99%

Robust Unsupervised Multifeature Representation for Infrared Small Target Detection

Chen,

Wu,

Zheng

et al. 2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Infrared small target detection is critical to infrared search and tracking (IRST) systems. However, accurate and robust detection remains challenging due to the scarcity of target information and the complexity of clutter interference. Existing methods have some limitations in feature representation, leading to poor detection performance in complex scenes. Especially when there are sharp edges near the target or in cluster multitarget detection, the "target suppression" phenomenon tends to occur. To address this issue, we propose a robust unsupervised multi-feature representation (RUMFR) method for infrared small target detection. On the one hand, robust unsupervised spatial clustering (RUSC) is designed to improve the accuracy of feature extraction; on the other hand, pixel-level multiple feature representation is proposed to fully utilize the target detail information. Specifically, we first propose the center-weighted interclass difference measure (CWIDM) with a trilayer design for fast candidate target extraction. Note that CWIDM also guides the parameter settings of RUSC. Then, the RUSC-based model is constructed to accurately extract target features in complex scenes. By designing the parameter adaptive strategy and iterative clustering strategy, RUSC can robustly segment cluster multi-targets from complex backgrounds. Finally, RUMFR that fuses pixel-level contrast, distribution and directional gradient features is proposed for better target representation and clutter suppression. Extensive experimental results show that our method has stronger feature representation capability and achieves better detection performance than several state-of-the-art methods.

show abstract

“…Intelligent monitoring of power M-LVSC involves the use of advanced technologies and tools to monitor and analyze the performance and condition of power distribution systems in real-time [3]. The goal of intelligent monitoring is to ensure the efficient,reliable,and safe operation of the power distribution system by identifying potential problems and taking corrective actions before they cause power outages or other disruptions [4]. The rapid development of intelligent technology has also promoted the construction of new power systems.…”

Section: Introductionmentioning

confidence: 99%

Multi-parameter fusion diagnosis for medium and lower voltage switchgear cabinet based on UHF and Infrared Camera Method

Li,

Gao,

Gong

et al. 2024

Preprint

View full text Add to dashboard Cite

Multi-parameter live detection and fusion diagnosis of medium and lower voltage switch-gear cabinet (M-LVSC) is an important technology for identification the operating status of power equipment. Under the operating conditions, a single sensor(UHF,thermal infrared,ultrasonic partial discharge detection) in the distribution equipment cannot judge the complex faults.The objective of this research is to develop fusion diagnosis and techniques for detect the state of 10kV and 400V distribution equipment in the switch-gear and its bus, identifying any abnormal conditions, by the infrared camera method and UHF partial discharge detection. The condition data of the equipment from these sensors is collected and analyzed to from the operating conditions by intelligent monitoring data technology for power distribution equipment.Under the background of big data cloud platform, overheat defects and discharge defects were captured and deeply analysis by UHF and infrared camera method. Fusion diagnosis provides an intelligent decision scheme for power distribution equipment fault diagnosis under operating conditions. The fusion recognition by comprehensive detect for a M-LVSC contactors power line and terminal board were verified in power distribution site.

show abstract

Prior-Guided Data Augmentation for Infrared Small Target Detection

Cited by 13 publications

References 54 publications

Generative Adversarial Differential Analysis for Infrared Small Target Detection

Generative Adversarial Differential Analysis for Infrared Small Target Detection

Robust Unsupervised Multifeature Representation for Infrared Small Target Detection

Multi-parameter fusion diagnosis for medium and lower voltage switchgear cabinet based on UHF and Infrared Camera Method

Contact Info

Product

Resources

About