Article
Geochemistry & Geophysics
Bing Yang, Hong Li, Ziyang Guo
Summary: This article proposes a structure-preserving spectral-spatial network (SPSSN) to extract discriminative deep structure-preserving spectral-spatial features by utilizing the manifold structure information during the feature learning process. Experiments on real HSI datasets verify the effectiveness and superiority of the SPSSN compared with several state-of-the-art methods in HSI classification.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Yuxuan Zheng, Jiaojiao Li, Yunsong Li, Jie Guo, Xianyun Wu, Yanzi Shi, Jocelyn Chanussot
Summary: This study proposes a novel edge-conditioned feature transform network, composed of three parts, to successfully address the challenges encountered in reconstructing high-spatial-resolution HSI, achieving more accurate spatial details and less spectral distortions.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Xinyu Zhang, Yantao Wei, Weijia Cao, Huang Yao, Jiangtao Peng, Yicong Zhou
Summary: The article proposes a spatial-spectral feature representation method based on local correntropy matrix for hyperspectral image (HSI) classification. By performing dimension reduction and constructing local correntropy matrix, the proposed method achieves competitive performance in HSI classification.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Nannan Liang, Puhong Duan, Haifeng Xu, Lin Cui
Summary: This paper proposes a multi-view structural feature extraction method that considers the correlation and dependencies of different regions to provide a complete characterization of the spectral-spatial structures of objects. Experimental results show that this method outperforms other state-of-the-art classification methods in terms of visual performance and objective results, especially with limited training set.
Article
Environmental Sciences
Aili Wang, Shuang Xing, Yan Zhao, Haibin Wu, Yuji Iwahori
Summary: This study proposes a method that combines a spectral spatial kernel and an improved Vision Transformer (ViT) for the classification task of hyperspectral images. By dimensionally reducing and extracting features from the hyperspectral data, and introducing a re-attention mechanism and a local mechanism, the method can better mine and represent the sequence properties of spectral features and utilize the local and global information of the data, thereby improving the classification accuracy.
Article
Environmental Sciences
Chunchao Li, Xuebin Tang, Lulu Shi, Yuanxi Peng, Yuhua Tang
Summary: This article proposes an efficient two-staged hyperspectral feature extraction method based on total variation. The method includes reducing the spectral dimension using the average fusion method in the first stage and obtaining featured blocks of different smoothness using the anisotropic total variation model in the second stage. The results demonstrate that the proposed method outperforms state-of-the-art methods in terms of classification accuracy and computing time, and it also shows robustness and stability through comprehensive parameter analysis.
Article
Geochemistry & Geophysics
Yifan Sun, Bing Liu, Xuchu Yu, Anzhu Yu, Kuiliang Gao, Lei Ding
Summary: In recent years, significant development has been achieved in deep-learning-based hyperspectral image (HSI) classification methods. This article proposes a framework called SMF-UL, which learns spectral variation knowledge through unsupervised learning on mass unlabeled HSI data to obtain a robust capability of feature extraction with generalization. Experimental results show that SMF-UL achieves competitive classification performance and demonstrates flexibility and superiority compared to advanced methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Xiangyong Cao, Jing Yao, Xueyang Fu, Haixia Bi, Danfeng Hong
Summary: The study proposes an enhanced 3DDWT approach to address noise contamination in hyperspectral image classification, utilizing a combination of CNN model and active learning strategy to achieve better categorized results compared to other advanced methods.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2021)
Article
Geochemistry & Geophysics
Chao Pan, Xiuping Jia, Jie Li, Xinbo Gao
Summary: This article introduces a novel approach for addressing the over-smoothing issue in MRF by class-by-class refinement and adaptive edge preservation. Experimental results demonstrate the superiority of aEPMs in evaluation metrics and detail preservation compared to traditional methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Geochemistry & Geophysics
Bing Zhang, Yashinov Aziz, Zhicheng Wang, Lina Zhuang, Michael K. Ng, Lianru Gao
Summary: The proposed GF-destriping algorithm efficiently detects and removes stripes in hyperspectral images using Gabor filters and FastHyIn inpainting method, outperforming state-of-the-art destriping algorithms in numerical experiments on simulated and real data sets.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Geochemistry & Geophysics
Jianmeng Li, Hui Sheng, Mingming Xu, Shanwei Liu, Zhe Zeng
Summary: This article presents a novel band-by-band adaptive multiscale superpixel feature extraction method (BAMS-FE) for hyperspectral imagery. It performs superpixel segmentation for each band and determines the optimal multiscale parameters using an unsupervised approach. The results demonstrate its excellent precision and stability.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Shuzhen Zhang, Ting Lu, Wei Fu, Shutao Li
Summary: A novel superpixel-level hybrid discriminant analysis (SHDA) method is proposed for hyperspectral image (HSI) classification. The method takes advantage of superpixels in characterizing spatial-spectral shape-adaptive structure and the power of discriminant analysis in enhancing class-separability. Two specific discriminant analysis modules, superpixel-level local discriminant analysis (SLDA) and superpixel-level nonlocal discriminant analysis (SNDA), are designed to effectively capture the local and nonlocal spatial-spectral correlation information. The proposed SHDA method outperforms several state-of-the-art techniques on three real hyperspectral datasets.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Qiaoqiao Sun, Xuefeng Liu, Salah Bourennane
Summary: A novel unsupervised multi-level feature extraction framework based on a three-dimensional convolutional autoencoder is proposed in this paper to improve hyperspectral classification. The framework allows for spectral-spatial information to be mined simultaneously and can be trained without labeled samples, providing more efficient feature extraction compared to using multiple networks.
Article
Computer Science, Artificial Intelligence
Wei Liu, Pingping Zhang, Yinjie Lei, Xiaolin Huang, Jie Yang, Michael Ng
Summary: In this paper, a generalized smoothing framework is introduced using the truncated Huber penalty function, which provides strong flexibility. The framework can be applied to various image processing tasks and achieves superior performance in challenging cases.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Engineering, Electrical & Electronic
Xiaochen Lu, Dezheng Yang, Fengde Jia, Yunlong Yang, Lei Zhang
Summary: The proposed multilevel joint feature extraction network has shown significant effectiveness in processing HSI classification tasks by extracting spectral-spatial features efficiently. Through a designed convolution process, information from each channel is transformed into valid channel-wised spatial features and forms global attention details to guide further feature mining. Ultimately, features obtained at different levels are integrated for ground object classification.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Computer Science, Information Systems
Haidong Wang, Xuan He, Zhiyong Li, Jin Yuan, Shutao Li
Summary: This study proposes an end-to-end MOT network called joint detection and association network (JDAN) that can simultaneously perform object detection and data association, resulting in improved tracking performance by optimizing the overall task.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Jinyang Liu, Renwei Dian, Shutao Li, Haibo Liu
Summary: This study proposes a saliency guided deep-learning framework for pixel-level image fusion, which can simultaneously deal with different tasks and generate fusion images that are more in line with visual perception by extracting meaningful information through fusion weights.
INFORMATION FUSION
(2023)
Article
Computer Science, Artificial Intelligence
Xiaohui Wei, Ting Lu, Shutao Li
Summary: Graphs are crucial for improving graph-based machine learning methods, but existing approaches have limitations when dealing with multiple graphs. This article presents a novel method that can simultaneously capture global and local information, and achieves superior results in clustering tasks compared to previous state-of-the-art methods.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Geochemistry & Geophysics
Zhuoyi Zhao, Xiang Xu, Jun Li, Shutao Li, Antonio Plaza
Summary: Nowadays, CNN-based DL models have gained popularity in HSIC and achieved high accuracy due to their hierarchical and nonlinear feature learning patterns. However, deeper network structures may demand more parameters and training samples. To overcome these problems, we propose a lightweight network model using the GSC module, which reduces parameters and is suitable for HSI data. Experimental results show that our model has low training cost and achieves competitive accuracy with fewer samples compared to existing models.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Ze Song, Xiaohui Wei, Xudong Kang, Shutao Li, Jinyang Liu
Summary: In this study, we propose a cross-temporal context learning network called CCLNet, which leverages intratemporal and intertemporal long-range dependencies to fully exploit cross-temporal context information. Our method achieves improved change detection performance, especially in complex and diverse changing scenes.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Computer Science, Artificial Intelligence
Anjing Guo, Renwei Dian, Shutao Li
Summary: In recent years, fusing a low-resolution hyperspectral image (LR-HSI) with a high-resolution multispectral image (HR-MSI) from different satellites has been proven to be an effective method for improving HSI resolution. However, the LR-HSI and HR-MSI obtained from different satellites may not satisfy existing observation models and it is difficult to register them. To address these issues, a deep-learning-based framework is proposed, which includes image registration, blur kernel learning, and image fusion. The proposed framework demonstrates superior performance in HSI registration and fusion accuracy through extensive experiments.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Zhiyong Lv, Haitao Huang, Weiwei Sun, Meng Jia, Jon Atli Benediktsson, Fengrui Chen
Summary: This article proposes an iterative training sample augmentation (ITSA) strategy to improve the performance of deep learning neural networks in land cover change detection (LCCD) tasks with remote sensing images. The strategy is verified with experiments on seven pairs of real remote sensing images, showing excellent visual performance and quantitative improvement in detection accuracies of LCCD. The results indicate that the deep learning network coupled with ITSA can effectively improve LCCD performance.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Haolong Fu, Shixun Wang, Puhong Duan, Changyan Xiao, Renwei Dian, Shutao Li, Zhiyong Li
Summary: Visible-infrared object detection aims to improve detector performance by fusing the complementarity of visible and infrared images. To overcome the limitation of existing methods that only utilize local intramodality information, we propose a feature-enhanced long-range attention fusion network (LRAF-Net) that leverages the long-range dependence between different modalities.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Geochemistry & Geophysics
Puhong Duan, Xudong Kang, Pedram Ghamisi, Shutao Li
Summary: Oil spill detection is a topic of increasing importance due to its negative impact on the environment and coastal communities. This study proposes an unsupervised method for oil spill detection using hyperspectral remote sensing images. The method utilizes isolation forest to estimate the probability of each pixel belonging to seawater or oil spills, and generates pseudolabeled training samples using clustering algorithms. The proposed method achieves superior detection performance compared to other state-of-the-art approaches.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Shuo Zhang, Puhong Duan, Xudong Kang, Yan Mo, Shutao Li
Summary: In this article, a feature-band-based unsupervised underwater target detection method (FBUD) is proposed, which uses the normalized difference water index (NDWI) and unmixing technique to find the target and background pixels. The spectral difference between the target and background is used to determine the optimal feature bands. The probability map of the underwater target can be obtained through a simple math operation on the feature bands. Additionally, a new UAV-borne hyperspectral image dataset named HNU-UTD is built for real-world underwater target detection, and the experimental results confirm the accuracy and effectiveness of the proposed method, which outperforms supervised detection methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Computer Science, Artificial Intelligence
Ze Song, Xudong Kang, Xiaohui Wei, Shutao Li
Summary: Camouflaged object detection is always faced with the challenge of identifying object pixels embedded in the background. Existing deep learning methods lack the ability to effectively utilize the context information around different pixels. In this paper, a pixel-centric context perception network (PCPNet) is proposed to address this problem. PCPNet customizes personalized context for each pixel based on the automatic estimation of its surroundings. Experimental results demonstrate the superiority of PCPNet in camouflaged object detection compared to other state-of-the-art methods.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Geochemistry & Geophysics
Huiling Gao, Shutao Li, Jun Li, Renwei Dian
Summary: This research proposes a dual-branch network with attention mechanisms for multispectral image pan-sharpening. By improving the linear transformation and decomposition methods, it enhances the performance of pan-sharpening and improves the physical interpretability.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Computer Science, Artificial Intelligence
Renwei Dian, Anjing Guo, Shutao Li
Summary: The paper introduces a zero-shot learning method for HSI sharpening, which estimates the spectral and spatial responses of imaging sensors and uses subsampled HSI and MSI for inference to improve sharpening performance. Additionally, dimension reduction is applied to the HSI to reduce the model size, and an imaging model-based loss function is designed to enhance fusion performance.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Haibo Liu, Chenguo Feng, Renwei Dian, Shutao Li
Summary: A spatial-spectral transformer-based U-net (SSTF-Unet) approach is proposed in this paper to achieve the fusion of high-resolution hyperspectral images and high-resolution multispectral images by capturing the association between distant features and exploring the intrinsic information of the images. The method utilizes spatial and spectral self-attention and incorporates multiple fusion blocks for multiscale feature fusion.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Qiya Song, Bin Sun, Shutao Li
Summary: Automatic speech recognition (ASR) is a crucial interface in intelligent systems, but its performance is often affected by external noise. Audio-visual speech recognition (AVSR) utilizes visual information to enhance ASR in noisy conditions. This article proposes a multimodal sparse transformer network (MMST) that incorporates sparse self-attention mechanism and motion features to improve AVSR performance.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)