Yuyang Wang

Yuyang Wang

Auto-generated profile, based on publicly available information

Yuyang Wang

Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 August 2025 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
ZSZhenqiu Shu
Zhenqiu Shu
KZKexin Zeng
Kexin Zeng
JZJun Zhou
Jun Zhou
YWYuyang Wang
Yuyang Wang
MTMingzhu Tai
Mingzhu Tai
ZYZhengtao Yu
Zhengtao Yu
The attention mechanism has gained significant popularity in hyperspectral image classification (HSIC) for its ability to adaptively highlight key features. However, it often incurs expensive computational costs due to the iterative operations. Moreover, the high inter-class similarity and indistinct class boundaries in hyperspectral images (HSIs) pose challenges for classification tasks. To tackle these issues, we propose a spectral-spatial synergy guided attention network (SSSGAN) for HSIC. It initially employs a spectral-spatial synergistic mechanism to jointly extract local spectral and spatial features of HSIs, and then applies a guided attention mechanism to effectively reduce computational complexity and further improve its discriminative capabilities. Specifically, a stacked residual spectral feature extractor stacks scale-variable convolution blocks within the channel-connected residual structure. It precisely localizes boundaries and reduces boundary blur by extracting and fusing local multi-level spectral features. A hierarchical dual feature enhancement module employs parallel multi-scale branches combined with dual shift operations to extract multi-scale spatially enhanced features, thereby enriching the diversity of feature representations and improving the discriminability of inter-class features. Finally, a feature-aware guided attention module reduces the computational cost of the attention mechanism by incorporating the guiding token and convolutional operations. By introducing local regularization into the attention map, the convolution enhances the robustness and effectiveness of global and contextual information extraction from HSIs. Experiments conducted on four publicly available benchmark datasets (PU, HU2013, SV, and LK) demonstrate that the proposed SSSGAN method achieves overall accuracies of 98.96%, 95.93%, 98.72%, and 98.31%, respectively, thereby validating its effectiveness in HISC tasks. The source code for this work can be found at https://github.com/szq0816/DSGAN.

28 October 2025 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
YWYuyang Wang
Yuyang Wang
MTMingzhu Tai
Mingzhu Tai
ZLZigao Liu
Zigao Liu
ZSZhenqiu Shu
Zhenqiu Shu
FXFengchao Xiong
Fengchao Xiong
ZYZhengtao Yu
Zhengtao Yu
Hyperspectral image (HSI) classification plays a major role in remote sensing. The combination of convolutional neural networks (CNNs) and Transformers has garnered increasing attention in HSI classification, yielding impressive performance. However, these approaches primarily focus on modeling local-global feature representations, often ignoring the rich multi-range information inherent in HSIs. To address these issues, this work proposes a Mid-range Convolutional Modulated Transformer Network (MCMTN) for HSI classification. Compared to conventional CNN and Transformer architectures, the proposed method employs large-kernel convolutions to simplify self-attention operations and facilitate the integration of short-, mid-, and long-range information in HSIs. Specifically, we employ a Lightweight Spatial-Spectral Module (LWSSM) to extract the short-range spatial features and spectral features of HSIs. Furthermore, we can effectively model the long-range spatial information in a Spatial Compression Module (SCM) using a simple compress-decompress strategy. We also introduce a Mid-range Convolutional Modulation (MCM) into the network, which effectively captures mid-range spatial dependencies in HSIs while enabling semantic information interaction in a Transformer manner. Compared with traditional HSI classification methods, our proposed MCMTN method can effectively extract multidimensional representations with less time consumption. The experimental results on several benchmark datasets show that the proposed MCMTN method achieves better classification performance than several state-of-the-art models. The code is available at https://github.com/szq0816/MCMTN.

14 April 2025 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
YWYuyang Wang
Yuyang Wang
ZSZhenqiu Shu
Zhenqiu Shu
ZYZhengtao Yu
Zhengtao Yu
Recently, Transformer has gained widespread application in hyperspectral image classification (HSIC) tasks due to its powerful global modeling ability. However, the inherent high-dimensional property of hyperspectral images (HSIs) leads to a sharp increase in the number of parameters and expensive computational costs. Moreover, self-attention operations in Transformer-based HSIC methods may introduce irrelevant spectral-spatial information, and thus may consequently impact the classification performance. To mitigate these issues, in this paper, we introduce an efficient deep network, named efficient attention Transformer network (EATN), for practice HSIC tasks. Specifically, we propose two self-similarity descriptors based on the original HSI patch to enhance spatial feature representations. The center self-similarity descriptor emphasizes pixels similar to the central pixel. In contrast, the neighborhood self-similarity descriptor explores the similarity relationship between each pixel and its neighboring pixels within the patch. Then we embed these two self-similarity descriptors into the original patch for subsequent feature extraction and classification. Furthermore, we design two efficient feature extraction modules based on the preprocessed patches, called spectral interactive Transformer (SIT) module and spatial conv-attention (SCA) module, to reduce the computational costs of the classification framework. Extensive experiments on four benchmark datasets show that our proposed EATN method outperforms other state-of-the-art HSI classification approaches. The source code of this work will be released later.

4 December 2025 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
ZSZhenqiu Shu
Zhenqiu Shu
KZKexin Zeng
Kexin Zeng
YWYuyang Wang
Yuyang Wang
STSongze Tang
Songze Tang
ZYZhengtao Yu
Zhengtao Yu
LXLiang Xiao
Liang Xiao
Transformer-based methods have recently shown remarkable success in hyperspectral image classification (HSIC). However, their applications, in practice, still face two significant challenges. First, although the multihead mechanism in self-attention improves model robustness during training, it may overlook the continuity of spectral bands. Second, existing methods often struggle to effectively balance global and local information during multiscale feature extraction, limiting further improvements in classification performance. To address these issues, we propose a novel spectral-guided multiscale feature-aware Transformer (SMFAT) framework for HSIC. Specifically, a global low-rank spectral learning (GLSL) module is introduced to project hyperspectral image patches into a low-rank subspace, reducing spectral redundancy and capturing global spectral correlations. Furthermore, we introduce the multiscale feature-aware self-attention (MFASA) mechanism, which dynamically integrates fine- and coarse-grained features to enhance multiscale feature modeling. Finally, a spectral-guided fusion (SGF) module leverages the global spectral information extracted by the GLSL module to guide MFASA in more effectively capturing interspectral correlations and spectral continuity. This approach facilitates a more effective integration of spectral and spatial features in HSIs. Experiments on three well-known HSI datasets verify that the proposed SMFAT method significantly outperforms several state-of-the-art approaches in real-world HSIC tasks. The source code for this work is available at https://github.com/stellaZ77/SMFAT