Article
Engineering, Aerospace
Boyang Nie, Lei Yang, Fei Zhao, Jinsong Zhou, Juanjuan Jing
Summary: A space object material identification method based on Tucker decomposition is proposed and demonstrated experimentally in this study, showing a superior performance compared to other decomposition methods.
ADVANCES IN SPACE RESEARCH
(2021)
Article
Chemistry, Analytical
Yiwei Mao, Christopher H. Betters, Samuel Garske, Jeremy Randle, K. C. Wong, Iver H. Cairns, Bradley J. Evans
Summary: This study successfully developed a fully open-source data acquisition system that can collect hyperspectral and navigation data concurrently for direct georeferencing. The system utilizes low-cost, lightweight, and deployable devices, and employs commercial-off-the-shelf hardware and open-source software, providing an affordable solution for hyperspectral data remote sensing.
Article
Ecology
Xiaoquan Pan, Jinbao Jiang, Yiming Xiao
Summary: Natural gas is an important clean energy source, but its leakage during transportation can have negative impacts. This study used hyperspectral remote sensing technology to indirectly detect natural gas leakage by analyzing the spectral characteristics of vegetation. The experiment found specific spectral bands sensitive to gas stress, which can be used to identify stressed plants. The proposed index showed promising results in identifying gas-stressed plants.
ECOLOGICAL INFORMATICS
(2022)
Article
Engineering, Environmental
Xiaolan Cai, Luyao Wu, Yunmei Li, Shaohua Lei, Jie Xu, Heng Lyu, Junda Li, Huaijing Wang, Xianzhang Dong, Yuxing Zhu, Gaolun Wang
Summary: Due to rapid urbanisation, urban water quality has been degraded by increased pollutants. A remote sensing identification method for urban water pollution sources, using unmanned aerial vehicle (UAV) hyperspectral images, was established. By analyzing fluorescent components and spectral indices, four types of pollution sources (domestic sewage, terrestrial input, agricultural and algal, and industrial wastewater) were identified. Optical parameters were used to develop an identification method with a recognition accuracy exceeding 70% for the four pollution sources, expanding the application of remote sensing technologies for urban water quality management.
JOURNAL OF HAZARDOUS MATERIALS
(2023)
Article
Environmental Sciences
Hao Guo, Wenxing Bao, Kewen Qu, Xuan Ma, Meng Cao
Summary: This paper proposes a new algorithm for multispectral and hyperspectral image fusion, which estimates high spatial resolution hyperspectral images using a coupled non-negative block-term tensor model and introduces total variation (TV). Experiments show that the performance of this method is better than the state-of-the-art methods.
Article
Mathematics
Lang Huyan, Ying Li, Dongmei Jiang, Yanning Zhang, Quan Zhou, Bo Li, Jiayuan Wei, Juanni Liu, Yi Zhang, Peng Wang, Hai Fang
Summary: In this study, a novel decomposed module called DecomResnet based on Tucker decomposition was proposed to deploy a CNN object detection model on a satellite. Tensors provide a natural and compact representation of CNN weights via suitable low-rank approximations. A remote sensing image object detection model compression framework based on low-rank decomposition was introduced, which achieved significant compression and speedup ratios with only slight decrease in mAP.
Article
Geochemistry & Geophysics
Yin-Ping Zhao, Hongyan Li, Yongyong Chen, Zhen Wang, Xuelong Li
Summary: In this article, the structured sparsity plus enhanced low rankness ((SELR)-E-2) method is proposed for hyperspectral anomaly detection (HAD). It adopts a weighted tensor Schatten-p norm and a structured sparse norm to represent the low-rank properties of the background and the sparsity of abnormal pixels, respectively. A position-based Laplace regularizer is also used to preserve the local structural details.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Remote Sensing
Yanfei Zhong, Xinyu Wang, Shaoyu Wang, Liangpei Zhang
Summary: This paper discusses the recent progress in Chinese spaceborne HRS, including typical satellite systems, data processing, and applications, as well as the future development trends of HRS in China.
GEO-SPATIAL INFORMATION SCIENCE
(2021)
Review
Environmental Sciences
Bowen Chen, Liqin Liu, Zhengxia Zou, Zhenwei Shi
Summary: This paper reviews representative methods for hyperspectral image target detection, and categorizes them into seven categories: hypothesis testing-based methods, spectral angle-based methods, signal decomposition-based methods, constrained energy minimization (CEM)-based methods, kernel-based methods, sparse representation-based methods, and deep learning-based methods. The basic principles, classical algorithms, advantages, limitations, and connections of these methods are comprehensively summarized, and critical comparisons are made on the summarized datasets and evaluation metrics. Furthermore, the future challenges and directions in the area are analyzed.
Article
Computer Science, Artificial Intelligence
Wei He, Yong Chen, Naoto Yokoya, Chao Li, Qibin Zhao
Summary: In this paper, a new coupled tensor ring factorization (CTRF) model is proposed for hyperspectral super-resolution (HSR). The CTRF model can effectively learn the tensor ring core tensors of high-resolution HSI while exploiting the low-rank property of each class, outperforming previous coupled tensor models.
PATTERN RECOGNITION
(2022)
Review
Engineering, Electrical & Electronic
Nour Aburaed, Mohammed Q. Alkhatib, Stephen Marshall, Jaime Zabalza, Hussain Al Ahmad
Summary: Remote sensing technology plays a crucial role in various industrial applications, and hyperspectral imaging (HSI) is an important modality. However, HSI suffers from low spatial resolution compared to multispectral images (MSI). Therefore, super resolution (SR) of HSI has gained attention in the past two decades. This article reviews important SR algorithms, datasets, sensors, and quality metrics in the field from 2002 to 2022, providing a summary of the current state and recommendations for future directions.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Geosciences, Multidisciplinary
Cheng Liu, Chengzhi Xing, Qihou Hu, Shanshan Wang, Shaohua Zhao, Meng Gao
Summary: This article reviews the recent advances in hyperspectral remote sensing techniques and discusses the future application prospects in air pollution monitoring. It recommends the use of a multi-means joint hyperspectral stereoscopic remote sensing monitoring mode for effective monitoring and regulation of air pollution.
EARTH-SCIENCE REVIEWS
(2022)
Review
Environmental Sciences
Alireza Sanaeifar, Ce Yang, Miguel de la Guardia, Wenkai Zhang, Xiaoli Li, Yong He
Summary: Recent advances and challenges in using proximal hyperspectral sensing for assessing plant abiotic stresses have been critically reviewed. This technique provides high-resolution images for studying plant physiology and monitoring spatio-temporal variations. The comprehensive review of 362 research papers shows the wide range of applications for detecting different types of abiotic stresses in plants.
SCIENCE OF THE TOTAL ENVIRONMENT
(2023)
Article
Chemistry, Analytical
Olivier Lim, Stephane Mancini, Mauro Dalla Mura
Summary: Hyperspectral imaging is a promising technique used in various fields. Compressive hyperspectral imaging devices, as an alternative to traditional devices, can reduce the number of acquisitions through compression. However, the reconstruction process is a limiting factor for the adoption of these devices due to its time-consuming nature and high computational burden. Algorithmic and material acceleration using embedded and parallel architectures can significantly speed up the image reconstruction process, making compressive hyperspectral systems suitable for real-time applications.
Article
Environmental Sciences
Le Sun, Qihao Cheng, Zhiguo Chen
Summary: The study proposed an HSI super-resolution model based on spectral smoothing prior and tensor tubal row-sparse representation, termed SSTSR, which reconstructs HSI with high spatial resolution and spectral resolution through nonlocal priors, tensor decomposition, and regularization. Experimental results showed that the method outperformed many advanced HSI super-resolution methods.