基于深度学习的高光谱图像分类方法.pdf

Artificial Intelligence and Robotics Research 人工智能与机器人研究, 2017, 6(1), 31-39 Published Online February 2017 in Hans. /journal/airr /10.12677/airr.2017.61005 A Classification Method for Hyperspectral Imagery Based on Deep Learning 1 1 2 3 3 Lin Yuan , Shaoxing Hu , Aiwu Zhang , Shatuo Chai , Xing Wang 1 School of Mechanical Engineering and Automation, Beihang University, Beijing 2 Colledge of Resource Environment and Tourism, Capital Normal University, Beijing 3 Animal husbandry and Veterinary Hospital of Qinghai University, Xining Qinghai rd th th Received: Feb. 3 , 2017; accepted: Feb. 18 , 2017; published: Feb. 24 , 2017 Abstract Remote sensing hyperspectral imaging can obtain abundant spectral information, which provides the possibility for the analysis of high precision terrain. The hyperspectral image has the charac- teristics of “map in one”, and the full use of spectral information and spatial information in hy- perspectral image is the premise of obtaining accurate classification results. Deep learning stack machine model in automatic encoding (Stack Auto-Encoder SAE) can effectively extract data in nonlinear information, and convolutional neural network (Convolutional Neural Network, CNN) can automatically extract features from the image. Based on this, this paper presents a classifica- tion method of hyperspectral images based on deep learning. Firstly, the spectral dimension of the hyperspectral data is reduced using automatic encoding machine, then convolutional neural net- work is used as the classifier, and the pixel and its neighborhood pixels a