Advances in Hyperspectral Image Processing Techniques

17Sparse Representation‐Based Hyperspectral Image Classification

Haoyang Yu¹, Jun Li², Wei Li³, and Bing Zhang⁴

¹Center for Hyperspectral Imaging in Remote Sensing (CHIRS), Information Science and Technology College, Dalian Maritime University, Dalian, China

²Guangdong Provincial Key Laboratory of Urbanization and Geo-Simulation, School of Geography and Planning, Sun Yat-sen University, Guangzhou, China

³School of Information and Electronics, Beijing Institute of Technology, Beijing, China

⁴Key Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, China

17.1 Introduction

Remote sensing (RS) is a technology of Earth observation, which has experienced a process from panchromatic image, multispectral image to hyperspectral image (HSI). HSI can be viewed as a data cube, which contains diagnostic continuous spectrum [1]. The syncretic form of image and spectrum brings HSI more details of terrestrial objects and benefits the precision land‐use and land‐cover mapping [2–4]. HSI classification (HSIC) uses classifier to determine the class of each pixel based on its features. Several classic models, such as support vector machine (SVM) and sparse representation (SR), have shown their availability for HSIC tasks [5–7].

One of the main problems of HSIC is the high correlation between adjacent bands and redundancy in both the spectral and spatial domains. Therefore, it has been demonstrated that HSI is essentially low rank ...