测绘通报 ›› 2019, Vol. 0 ›› Issue (2): 99-102,136.doi: 10.13474/j.cnki.11-2246.2019.0052

• 技术交流 • 上一篇    下一篇

深度学习在遥感影像分类与识别中的研究进展综述

王斌1,2, 范冬林1,2   

  1. 1. 桂林理工大学测绘地理信息学院, 广西 桂林 541004;
    2. 广西空间信息与测绘重点实验室, 广西 桂林 541004
  • 收稿日期:2018-05-06 修回日期:2018-06-21 出版日期:2019-02-25 发布日期:2019-03-05
  • 通讯作者: 范冬林。E-mail:dlfan@glut.edu.cn E-mail:dlfan@glut.edu.cn
  • 作者简介:王斌(1996-),男,主要研究方向为GIS和遥感图像处理。E-mail:bz21422@163.com
  • 基金资助:

    广西空间信息与测绘重点实验室基金(16-380-25-24;163802515;151400720);广西"八桂学者"岗位专项经费资助项目

Research progress of deep learning in classification and recognition of remote sensing images

WANG Bin1,2, FAN Donglin1,2   

  1. 1. College of Geomatics and Geoinformation, Guilin University of Technology, Guilin 541004, China;
    2. Guangxi Key Laboratory of Spatial Information and Geomatics, Guilin 541004, China
  • Received:2018-05-06 Revised:2018-06-21 Online:2019-02-25 Published:2019-03-05

摘要:

深度学习一直是机器学习和人工智能研究的热门主题,特别是将深度学习这一深层网络学习算法和遥感影像分类与识别联合起来,使得传统训练算法的局部最小性得以解决。本文首先简要介绍了遥感影像分类与识别算法的发展和经典算法的局限性,其次介绍了深度学习的几种主流算法并分析它们在遥感影像分类与识别处理方面的应用现状,最后对未来深度学习应用于遥感识别与分类趋势进行了展望。

关键词: 深度学习, 分类与识别, 遥感影像, 应用现状

Abstract:

The concept of deep learning has been a hot topic in machine learning and artificial intelligence since it was put forward. In particular, the deep learning, a deep neural network learning machine algorithm, is closely combined with the classification and recognition of remote sensing images, so that the local minimization of the traditional training algorithm can be solved. This paper briefly introduces the development of remote sensing image classification and recognition algorithms and the limitations of classical algorithms. Secondly, it introduces several mainstream algorithms of deep learning and analyzes their application status in remote sensing image classification and recognition processing. Finally, the future trend of application of deep learning in remote sensing recognition and classification is prospected.

Key words: deep learning, classification and recognition, remote sensing image, application status

中图分类号: