光学学报, 2019, 39 (5): 0528004, 网络出版: 2019-05-10
基于多特征和改进稀疏表示的高光谱图像分类 下载: 1197次
Hyperspectral Image Classification via Multiple-Feature-Based Improved Sparse Representation
遥感 高光谱图像 稀疏表示 特征提取 Gabor滤波 局部二值模式 remote sensing hyperspectral image sparse representation feature extraction Gabor filter local binary pattern (LBP)
摘要
为了实现对高光谱图像的分类,提出了一种基于多特征和改进稀疏表示的方法——MFISR。从高光谱图像中提取光谱特征、Gabor特征和局部二值模式(LBP)特征,求解稀疏系数,同时增加一个2范式约束,利用所得系数得到每个测试像素的最终类别标签。实验结果表明:所提MFISR方法对小样本的检测效果显著,分类性能稳定且较优。
Abstract
A multiple-feature-based improved sparse representation (MFISR) method is proposed herein for the classification of hyperspectral images. The spectral feature, Gabor feature, and local binary pattern (LBP) feature are extracted from the hyperspectral image; subsequently, the sparse coefficients are solved and a 2-paradigm constraint is added. These obtained coefficients are used to determine the final class label of each test pixel. The experimental results demonstrate that the proposed MSIFR method exhibits excellent results for the detection of small samples, and its classification performance is stable and good.
李非燕, 霍宏涛, 李静, 白杰. 基于多特征和改进稀疏表示的高光谱图像分类[J]. 光学学报, 2019, 39(5): 0528004. Feiyan Li, Hongtao Huo, Jing Li, Jie Bai. Hyperspectral Image Classification via Multiple-Feature-Based Improved Sparse Representation[J]. Acta Optica Sinica, 2019, 39(5): 0528004.