高光谱成像
计算机科学
模式识别(心理学)
人工智能
特征提取
特征选择
降维
特征(语言学)
卷积(计算机科学)
维数之咒
人工神经网络
数据挖掘
哲学
语言学
作者
Zhixi Feng,Xuehu Liu,Shuyuan Yang,Kai Zhang,Licheng Jiao
出处
期刊:IEEE Geoscience and Remote Sensing Letters
[Institute of Electrical and Electronics Engineers]
日期:2023-01-01
卷期号:20: 1-5
被引量:5
标识
DOI:10.1109/lgrs.2023.3236672
摘要
Most existing classification methods design complicated and large deep neural network (DNN) model to deal with the ubiquitous spectral variability and nonlinearity of hyperspectral images (HSIs). However, their application is blocked by limited training samples and considerable computational costs in real scenes. To solve these problems, we propose a simple spectral hierarchical feature fusion and selection network (HFFSNet). Specifically, we apply 1-D grouped convolution for dimensionality reduction and multilevel feature extraction, then the multilevel features are fused to assist the adaptive feature selection of different layer features via the soft attention mechanism, and finally the selected features are fused to further enhance the feature representation. Extensive experimental results on three hyperspectral datasets demonstrate the effectiveness of the proposed network.
科研通智能强力驱动
Strongly Powered by AbleSci AI