数字农科院2.0

BTSM-UNet: Bidirectional Temporal-Spatial vision Mamba-UNet for winter wheat classification based on Sentinel-2 imagery

文献类型： 外文期刊

作者： Fan, Lingling;Chen, Shi;Xia, Lang;Zha, Yan;Zhou, Qingbo;Yan, Qi

作者机构：

关键词： Deep learning;Wheat classification;Multi-temporal imagery;State-space sequence model;Feature fusion

期刊名称： SMART AGRICULTURAL TECHNOLOGY

ISSN：

年卷期： 2025 年 12 卷

页码：

收录情况： 无来源刊（2025版）

摘要： In recent years, deep learning has made significant advancements in the field of semantic segmentation of remote sensing images, owing to its exceptional feature extraction capabilities and robust modeling performance. However, due to the high dimensionality and multi-channel characteristics of time series remote sensing images, most current semantic segmentation algorithms still find it difficult to automatically and efficiently extract image features. Moreover, most models often overlook the integration of global and local information at multiple scales, especially for time series remote sensing images with large-scale variations. In this study, we propose a novel Bidirectional Temporal-Spatial vision Mamba-UNet (BTSM-UNet) for winter wheat mapping based on state space sequence models, which can efficiently capture complex dependencies in sequential data. This model follows an Encoder + Decoder structure, and a Bidirectional Temporal-Spatial vision Mamba (Bi-TSM) module is inserted to fully exploit spatiotemporal information from time series imagery. The results indicate that our BTSM-UNet achieved the best classification accuracy with an MCC of 0.905 and an F1-score of 0.952 for winter wheat mapping in the North China Plain. Additionally, the results of band combinations demonstrated that our proposed model exhibited better stability compared to other deep models. Compared with the results of all band combinations, we found that the MCC (MCC/All) of BTSM-UNet decreased by only 2.43% when using RGB bands, while it decreased by 28.4% for the CNN-BiLSTM model, further proving the effectiveness and generalization ability of our method. Moreover, the BTSM-UNet exhibited the fastest data processing rate (24 FPS), a relatively small number of model parameters (4.86 M), and low computational complexity (60.19 G) compared to other models. We expect that the proposed method will provide a practical and effective technical approach for accurately and intelligently mapping winter wheat fields.

分类号：

  • 相关文献

[1]Intelligent identification on cotton verticillium wilt based on spectral and image feature fusion. Zhihao Lu,Shihao Huang,Xiaojun Zhang,Yuxuan shi,Wanneng Yang,Longfu Zhu,Chenglong Huang. 2023

[2]Diagnosing Cotton Farmland Quality Using Multi-Temporal Remotely Sensed Data. Bai, Junhua,Li, Shaokun,Bai, Junhua,Wang, Xu,Bai, Junhua,Li, Jing,Liu, Qinhuo,Bai, Junhua,Li, Jing,Liu, Qinhuo.

[3]Noise-tolerant RGB-D feature fusion network for outdoor fruit detection. Qixin Sun,Xiujuan Chai,Zhikang Zeng,Guomin Zhou,Tan Sun. 2022

[4]Exploring Multisource Feature Fusion and Stacking Ensemble Learning for Accurate Estimation of Maize Chlorophyll Content Using Unmanned Aerial Vehicle Remote Sensing. Weiguang Zhai,Changchun Li,Qian Cheng,Fan Ding,Zhen Chen. 2023

[5]Multimodal fusion-based detection method of estrus cows using multisource data inspired by hidden Markov model algorithms. Jun Wang,Yijia Zhao,Xiaoxia Li,Yu Zhou,Kaixuan Zhao,Hui Wang,Waleid Mohamed EL-Sayed Shakweer. 2025

[6]Field Rice Growth Monitoring and Fertilization Management Based on UAV Spectral and Deep Image Feature Fusion. Chen, Bingnan,Su, Qihe,Li, Yansong,Chen, Rui,Yang, Wanneng,Huang, Chenglong. 2025

[7]An Improved Model for Online Detection of Early Lameness in Dairy Cows Using Wearable Sensors: Towards Enhanced Efficiency and Practical Implementation. Xiaofei Dai,Guodong Cheng,Lu Yang,Yali Wang,Zhongkun Li,Shuqing Han,Jifang Liu. 2025

[8]The Bayesian mixture expert recognition model for tobacco leaf curing stages based on feature fusion. Panzhen Zhao,Shijiang Duan,Songfeng Wang,Aihua Wang,Lingfeng Meng,Zhicheng Wang,Yingpeng Dai. 2025

[9]Detection of Fusarium Head Blight in Individual Wheat Spikes Using Monocular Depth Estimation with Depth Anything V2. Jiacheng Wang,Jianliang Wang,Yuanyuan Zhao,Fei Wu,Wei Wu,Zhen Li,Chengming Sun,Tao Li,Tao Liu. 2025

[10]Research on the Classification Method of Fresh Tobacco Leaf Maturity Based on Transfer Learning and Multi-Feature Fusion. Zhao, Panzhen,Dai, Yingpeng. 2025

[11]Deep learning based weed detection and target spraying robot system at seedling stage of cotton field. Xiangpeng Fan,Xiujuan Chai,Jianping Zhou,Tan Sun. 2023

[12]Inter-Continental Transfer of Pre-Trained Deep Learning Rice Mapping Model and Its Generalization Ability. Lingbo Yang,Ran Huang,Jingcheng Zhang,Jingfeng Huang,Limin Wang,Jiancong Dong,Jie Shao. 2023

[13]Semantic Segmentation Based on Temporal Features: Learning of Temporal-Spatial Information From Time-Series SAR Images for Paddy Rice Mapping. Yang, Lingbo,Huang, Ran,Huang, Jingfeng,Lin, Tao,Wang, Limin,Mijiti, Ruzemaimaiti,Wei, Pengliang,Tang, Chao,Shao, Jie,Li, Qiangzi,Du, Xin. 2021

[14]A Novel Physics-Statistical Coupled Paradigm for Retrieving Integrated Water Vapor Content Based on Artificial Intelligence. Ruyu Mei,Kebiao Mao,Jiancheng Shi,Jeffrey Nielson,Sayed M. Bateni,Fei Meng,Guoming Du. 2023

[15]Machine learning for image-based multi-omics analysis of leaf veins. Yubin Zhang,Ning Zhang,Xiujuan Chai,Tan Sun. 2023

[16]Auction-based deep learning-driven smart agricultural supply chain mechanism. Yu Feng,Dong Mei,Hua Zhao. 2023

[17]Research on machine vision and deep learning based recognition of cotton seedling aphid infestation level. Xin Xu,Jing Shi,Yongqin Chen,Qiang He,Liangliang Liu,Tong Sun,Ruifeng Ding,Yanhui Lu,Chaoqun Xue,Hongbo Qiao. 2023

[18]Integrating satellite-derived climatic and vegetation indices to predict smallholder maize yield using deep learning. Liangliang Zhang,Zhao Zhang,Yuchuan Luo,Juan Cao,Ruizhi Xie,Shaokun Li. 2021

[19]A rapid, low-cost deep learning system to classify strawberry disease based on cloud service. Guo feng YANG,Yong YANG,Zi kang HE,Xin yu ZHANG,Yong HE. 2022

[20]Deep learning for change detection in remote sensing: a review. Bai, Ting,Wang, Le,Yin, Dameng,Sun, Kaimin,Chen, Yepei,Li, Wenzhuo,Li, Deren. 2022

作者其他论文 更多>>