HyperSFormer: A Transformer-Based End-to-End Hyperspectral Image Classification Method for Crop Classification
Remote Sensing · 11.07.2023 · 15 (14) · 3491
DOI: 10.3390/rs15143491 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 11.07.2023. Verfasst von Jiaxing Xie und 9 weiteren Autor:innen. Die Arbeit wurde bisher 36-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Laboratory of Lingnan Modern Agriculture 713 Science and Technology Guangdong Experimental Heyuan Branch Project 714, Co-constructing Cooperative Project on Agricultural Sci-tech of New Rural Development Research Institute of South China Agricultural University, the Guangdong Provincial Special Fund for Modern Agriculture Industry Technology Innovation Teams, China.
Abstract (Originalsprache)
Crop classification of large-scale agricultural land is crucial for crop monitoring and yield estimation. Hyperspectral image classification has proven to be an effective method for this task. Most current popular hyperspectral image classification methods are based on image classification, specifically on convolutional neural networks (CNNs) and recurrent neural networks (RNNs). In contrast, this paper focuses on methods based on semantic segmentation and proposes a new transformer-based approach called HyperSFormer for crop hyperspectral image classification. The key enhancement of the proposed method is the replacement of the encoder in SegFormer with an improved Swin Transformer while keeping the SegFormer decoder. The entire model adopts a simple and uniform transformer architecture. Additionally, the paper introduces the hyper patch embedding (HPE) module to extract spectral and local spatial information from the hyperspectral images, which enhances the effectiveness of the features used as input for the model. To ensure detailed model processing and achieve end-to-end hyperspectral image classification, the transpose padding upsample (TPU) module is proposed for the model’s output. In order to address the problem of insufficient and imbalanced samples in hyperspectral image classification, the paper designs an adaptive min log sampling (AMLS) strategy and a loss function that incorporates dice loss and focal loss to assist model training. Experimental results using three public hyperspectral image datasets demonstrate the strong performance of HyperSFormer, particularly in the presence of imbalanced sample data, complex negative samples, and mixed sample classes. HyperSFormer outperforms state-of-the-art methods, including fast patch-free global learning (FPGA), a spectral–spatial-dependent global learning framework (SSDGL), and SegFormer, by at least 2.7% in the mean intersection over union (mIoU). It also improves the overall accuracy and average accuracy values by at least 0.9% and 0.3%, respectively, and the kappa coefficient by at least 0.011. Furthermore, ablation experiments were conducted to determine the optimal hyperparameter and loss function settings for the proposed method, validating the rationality of these settings and the fusion loss function.
Förderung
- Laboratory of Lingnan Modern Agriculture 713 Science and Technology Guangdong Experimental Heyuan Branch Project 714
- Co-constructing Cooperative Project on Agricultural Sci-tech of New Rural Development Research Institute of South China Agricultural University
- the Guangdong Provincial Special Fund for Modern Agriculture Industry Technology Innovation Teams, China
- National Natural Science Foundation of China
- the Guangdong Science and Technology Innovation Cultivation Special Fund Project for College Students (“Climbing Program” Special Fund), China
Zitieren
APA
Xie, J., Hua, J., Chen, S., Wu, P., Gao, P., Sun, D., Lyu, Z., Lyu, S., Xue, X., & Lu, J. (2023). HyperSFormer: A Transformer-Based End-to-End Hyperspectral Image Classification Method for Crop Classification. Remote Sensing, 15(14), 3491. https://doi.org/10.3390/rs15143491
Vancouver
Xie J, Hua J, Chen S, Wu P, Gao P, Sun D, et al. HyperSFormer: A Transformer-Based End-to-End Hyperspectral Image Classification Method for Crop Classification. Remote Sensing. 2023;15(14):3491. doi:10.3390/rs15143491
BibTeX
@article{Xie2023,
author = {Xie, Jiaxing and Hua, Jiajun and Chen, Shaonan and Wu, Peiwen and Gao, Peng and Sun, Daozong and Lyu, Zhendong and Lyu, Shilei and Xue, Xiuyun and Lu, Jianqiang},
title = {HyperSFormer: A Transformer-Based End-to-End Hyperspectral Image Classification Method for Crop Classification},
journal = {Remote Sensing},
year = {2023},
volume = {15},
number = {14},
pages = {3491},
doi = {10.3390/rs15143491}
}
EUVIMED