A Distributed and Parallel Anomaly Detection in Hyperspectral Images Based on Low-Rank and Sparse Representation

Li, Jun; Weixuan, Zhang; Wu, Zebin; Zhang, Yi; Xu, Yang; Qian, Ling; Wei, Zhihui

doi:10.1109/igarss.2018.8517395

Cited by 3 publications

(4 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are also detection algorithms based on sparse representation [15,16], which represent the image background as a more representative basis vector or spectrum and use the product of spectral prior knowledge and related parameters to represent the original hyperspectral data. Li et al proposed the BJSR (background joint sparse representation) algorithm [17], an anomaly detection algorithm for hyperspectral images using background joint sparse representation by estimating an adaptive orthogonal background complementary subspace by BJSR, which adaptively selects the most representative background basis vectors for local regions, and then proposed an unsupervised adaptive subspace detection method to suppress the background and highlight the anomalous components at the same time.…”

Section: Related Wordmentioning

confidence: 99%

“…Here, ClassNum is the number of models that can be classified, and 3 is the three prediction frames that the model predicts for each pixel. The target coordinates lose the predicted output sizes for (52, 52, 15), (26,26,15) and (13,13,15). 15 is calculated from the coordinate point and confidence from the three boxes, whereas the confidence level is used to determine if there is a target in the box.…”

Section: Sehyp Headmentioning

confidence: 99%

“…The first two lines of the loss function are the loss function for the predicted frame. The CIOU algorithm [15,31] was used.…”

Section: Loss Functionmentioning

confidence: 99%

See 2 more Smart Citations

Intelligent hyperspectral target detection for reliable IoV applications

Wang

Jin

2022

J Wireless Com Network

View full text Add to dashboard Cite

In recent years, hyperspectral imagery has played a significant role in IoV (Internet of Vehicles) vision areas such as target acquisition. Researchers are focusing on integrating detection sensors, detection computing units, and communication units into vehicles to expand the scope of target detection technology with hyperspectral imagery. As imaging spectroscopy technology gradually matures, the spectral resolution of captured hyperspectral images is increasing. At the same time, the volume of data is also increasing. As a result, the reliability of IoV applications is challenged. In this paper, an intelligent hyperspectral target detection method based on deep learning network is proposed. It is based on the residual network structure with the addition of an attention mechanism. The trained network model requires few computational resources and can provide the results in a short time. Our method improves the value of mAP50 by an average of 3.57% for all categories and by up to 5% for a single category on the public dataset.

show abstract

Section: Related Wordmentioning

confidence: 99%

Section: Sehyp Headmentioning

confidence: 99%

See 1 more Smart Citation

Intelligent hyperspectral target detection for reliable IoV applications

Wang

Jin

2022

J Wireless Com Network

View full text Add to dashboard Cite

show abstract

“…Ren capitalized on the characteristic that the high-order automatic anomaly detection algorithm may descend into a local extreme point after randomly selecting the initial vector, facilitating parallel detection under diverse situations [30]. Liu optimized the target detection algorithm based on low-rank sparse representation [31]. They first segmented the image using the narrow dependency of hyperspectral data on the high-performance computing platform Spark, then used parallel clustering algorithm to cluster the pixels of hyperspectral images, and finally computed the clustered data in parallel, significantly improving the speed and scalability of the method.…”

Section: Introductionmentioning

confidence: 99%

A Parallel Algorithm for Hyperspectral Target Detection Based on Weighted Alternating Direction Method of Multiplier

Yu,

Wu,

et al. 2023

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

Self Cite

View full text Add to dashboard Cite

Target detection for hyperspectral images (HSIs) is one of the significant techniques in remote sensing data processing. Targets generally comprise various object categories with complex features and of varying sizes. Target detection is often used in complex application scenarios in which accurately and efficiently acquiring detection results can be challenging. The development of advanced target detection approaches is becoming increasingly necessary in both military and civilian fields. This article proposes an alternating direction method of multiplier (ADMM) based parallel approach for hyperspectral target detection. Different from existing methods performing target detection solely on hyperspectral images, our approach performs the fusion of hyperspectral and multispectral data to leverage both spectral and spatial information prior. For each task or data partition, the parallel processing of the computation load on multiple computing nodes can substantially reduce the computation time. In addition, we introduce a novel weighted ADMM, which takes the influence of different variables on convergence into account, to further enhance the computational efficiency of the target detection model. Experiments on real-world HSI datasets demonstrate that our proposed parallel method not only produces more accurate detection results than direct detection methods, but also achieves significant acceleration ratio compared with the serial processing flow.

show abstract