Hyperspectral and lidar data land-use classification using parallel transformers
文献类型:会议论文
作者 | Yuxuan, Hu1,2; Hao, He1,2; Lubin, Weng2 |
出版日期 | 2022-07 |
会议日期 | 2022-7-17 -> 2022-7-22 |
会议地点 | 线上会议 |
关键词 | Hyperspectral LiDAR Data Fusion Transformer Crossmodal |
英文摘要 | It has been proved that the fusion of hyperspectral and LiDAR data can effectively improve the performance of land-use classification. Hyperspectral data contain more information than LiDAR data but most recent models pay more attention to the design of feature fusion mechanisms. They use CNN which is not powerful enough in extracting spatialspectral features of hyperspectral data. In this paper, a simple yet effective model with parallel transformers is proposed. Transformer is a powerful tool for both feature extraction and feature fusion. One transformer acts as an hyperspectral image feature extractor while the other transformer is responsible for capturing crossmodal interactions. Experiments on Houston dataset and MUUFL Gulfport dataset demonstrate that the proposed model has significantly better performance than other state-of-the-art models. |
源URL | [http://ir.ia.ac.cn/handle/173211/48691] |
专题 | 自动化研究所_模式识别国家重点实验室_遥感图像处理团队 |
通讯作者 | Lubin, Weng |
作者单位 | 1.School of Artificial Intelligence, University of Chinese Academy of Sciences 2.National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences |
推荐引用方式 GB/T 7714 | Yuxuan, Hu,Hao, He,Lubin, Weng. Hyperspectral and lidar data land-use classification using parallel transformers[C]. 见:. 线上会议. 2022-7-17 -> 2022-7-22. |
入库方式: OAI收割
来源:自动化研究所
其他版本
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。