数字农科院2.0

Multi-Temporal Hyperspectral Classification of Grassland Using Transformer Network

文献类型： 外文期刊

作者： Xuanhe Zhao;Shengwei Zhang;Ruifeng Shi;Weihong Yan;Xin Pan

作者机构：

关键词： grassland;hyperspectral classification;multi-temporal;transformer network

期刊名称： Sensors

ISSN： 1424-8220

年卷期： 2023 年 23 卷 14 期

页码：

收录情况： SCIE（2023版） ; ; EI（2023版）

摘要： In recent years, grassland monitoring has shifted from traditional field surveys to remote-sensing-based methods, but the desired level of accuracy has not yet been obtained. Multi-temporal hyperspectral data contain valuable information about species and growth season differences, making it a promising tool for grassland classification. Transformer networks can directly extract long-sequence features, which is superior to other commonly used analysis methods. This study aims to explore the transformer network’s potential in the field of multi-temporal hyperspectral data by fine-tuning it and introducing it into high-powered grassland detection tasks. Subsequently, the multi-temporal hyperspectral classification of grassland samples using the transformer network (MHCgT) is proposed. To begin, a total of 16,800 multi-temporal hyperspectral data were collected from grassland samples at different growth stages over several years using a hyperspectral imager in the wavelength range of 400–1000 nm. Second, the MHCgT network was established, with a hierarchical architecture, which generates a multi-resolution representation that is beneficial for grass hyperspectral time series’ classification. The MHCgT employs a multi-head self-attention mechanism to extract features, avoiding information loss. Finally, an ablation study of MHCgT and comparative experiments with state-of-the-art methods were conducted. The results showed that the proposed framework achieved a high accuracy rate of 98.51% in identifying grassland multi-temporal hyperspectral which outperformed CNN, LSTM-RNN, SVM, RF, and DT by 6.42–26.23%. Moreover, the average classification accuracy of each species was above 95%, and the August mature period was easier to identify than the June growth stage. Overall, the proposed MHCgT framework shows great potential for precisely identifying multi-temporal hyperspectral species and has significant applications in sustainable grassland management and species diversity assessment.

分类号：

  • 相关文献

[1]Multi-temporal grass hyperspectral classification via full pixel decomposition spectral manifold projection and boosting active learning model. Zhao X.,Zhang S.,Yan W.,Pan X.. 2023

[2]2014年9月草业科学大事记. 张秀敏,刘阳. 2014

[3]2014年6月草业科学大事记. 张秀敏,刘阳. 2014

[4]不同改良措施对大针茅植物群落影响的研究. 闫志坚. 2002

[5]Rice identification using TerraSAR-X data. Guo, Lin,Pei, Zhiyuan,Zhang, Songling,Wang, Qingfa,McNairn, Heather,Shang, Jiali,Jiao, Xianfeng. 2011

[6]Crop Model Data Assimilation with Particle Filter for Yield Prediction Using Leaf Area Index of Different Temporal Scales. Li, He,Chen, Zhongxin,Wu, Wenbin,Liu, Bin,Hasi, Tuya,Li, He,Chen, Zhongxin,Wu, Wenbin,Liu, Bin,Hasi, Tuya. 2015

[7]The effects of Landsat image acquisition date on winter wheat classification in the North China Plain. Lingling Fan,Jing Yang,Xiao Sun,Fen Zhao,Shefang Liang,Dingding Duan,Hao Chen,Lang Xia,Jing Sun,Peng Yang. 2022

[8]Feature-Ensemble-Based Crop Mapping for Multi-Temporal Sentinel-2 Data Using Oversampling Algorithms and Gray Wolf Optimizer Support Vector Machine. Zhang H.,Gao M.,Ren C.. 2022

[9]Mangrove tree height growth monitoring from multi-temporal UAV-LiDAR. Yin D.,Wang L.,Lu Y.,Shi C.. 2024

[10]UAV imaging hyperspectral for barnyard identification and spatial distribution in paddy fields. Yanchao Zhang,Ziyi Yan,Junfeng Gao,Yiyang Shen,Haozhe Zhou,Wei Tang,Yongliang Lu,Yongjie Yang. 2024

[11]Applicability study of using in-situ gamma-ray spectrometry technique for Cs-137 and Pb-210(ex) inventories measurement in grassland environments. Li, Junjie,Wang, Yanglin,Wu, Jiansheng,Li, Junjie,Wang, Yanglin,Wu, Jiansheng,Li, Junjie,Li, Yong. 2010

[12]Establishing the carrying capacity of the grasslands of China: a review. Zhang, Y. J.,Zhang, X. Q.,Wang, X. Y.,Liu, N.,Kan, H. M.,Zhang, X. Q.. 2014

[13]Changes in soil carbon stocks and related soil properties along a 50-year grassland-to-cropland conversion chronosequence in an agro-pastoral ecotone of Inner Mongolia, China. Jiao, Yan,Zhao, JiaoHong,Yang, WenZhu,Jiao, Yan,Xu, Zhu. 2012

[14]Changes in Global Grassland Productivity during 1982 to 2011 Attributable to Climatic Factors. Gao, Qingzhu,Wan, Yunfan,Qin, Xiaobo,Ma, Xin,Liu, Shuo,Li, Yue,Gao, Qingzhu,Wan, Yunfan,Qin, Xiaobo,Ma, Xin,Liu, Shuo,Li, Yue,Schwartz, Mark W.,Williamson, Matthew A.,Peters, Casey B.,Zhu, Wenquan. 2016

[15]Quantitative Analysis of Plant Consumption and Preference by Oedaleus asiaticus (Acrididae: Oedipodinae) in Changed Plant Communities Consisting of Three Grass Species. Zhang, Z.,McNeill, M..

[16]Temporal variations in PM(10) and particle size distribution during Asian dust storms in Inner Mongolia. Hoffmann, Carsten,Funk, Roger,Sommer, Michael,Li, Yong. 2008

[17]Drivers of nitrous oxide fluxes from the semi-arid Leymus chinensis grassland in Inner Mongolia, China. Liu, Xingren,Zhang, Qingzhong,Liu, Xingren,Dong, Yunshe,Ren, Jianqiang. 2012

[18]Simulation of nitrous oxide emissions at field scale using the SPACSYS model. Wu, L.,Zhang, Xubo,Rees, R. M.,Tarsitano, D.,Jones, S. K.,Zhang, Xubo,Whitmore, A. P..

[19]Management opportunities to mitigate greenhouse gas emissions from Chinese agriculture. Nayak, Dali,Smith, Pete,Saetnan, Eli,Newbold, Jamie,Cheng, Kun,Cheng, Yan-Fen,Zhu, Wei Yun,Pan, Genxing,Wang, Wen,Wang, Wen,Koslowski, Frank,Moran, Dominic,Wang, Jia-Kun,Liu, Jian-Xin,Yan, Xiaoyuan,Cardenas, Laura,Lui, Yuelai.

[20]Analysis of land use change and its driving force in the Longitudinal Range-Gorge Region. Ll ZhengHai,Song GuoBao,Bao YaJing,Song GuoBao,Lo HaiYan,Ll ZhengHai,Wang HaiMei,Xu Tian,Cheng Yan.

作者其他论文 更多>>