首页  >  科研动态  >  正文
科研动态
窦鹏、李志伟的论文在IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 刊出
发布时间:2021-04-15 11:15:10     发布者:易真     浏览次数:

标题: Remote Sensing Image Classification Using Deep-Shallow Learning

作者: Dou, P (Dou, Peng); Shen, HF (Shen, Huanfeng); Li, ZW (Li, Zhiwei); Guan, XB (Guan, Xiaobin); Huang, WL (Huang, Wenli)

来源出版物: IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING  : 14  : 3070-3083  DOI: 10.1109/JSTARS.2021.3062635  出版年: 2021  

摘要: Recently, classification using multiple classifier system (MCS) has been reported as an effective method to improve remote sensing (RS) image classification. Such systems provide a complementary mechanism to use multiple classifiers, which have shallow architecture to solve the same classification problem; however, the system exhibits shortcomings due to complex ensemble strategy. Deep learning (DL) has been proven to be an advanced method for complex data classification; however, how to use its advantages to overcome the shortcomings of MCS in ensemble strategy for classification accuracy improvement is worthy of study. Thus, with the multiple classifier mechanism and DL architecture, we propose a novel RS image classification framework, namely, deep-shallow learning (DSL), to improve classification accuracy. The DSL framework consists of a shallow learning (SL) layer and a DL layer. The SL layer contains various classifiers with shallow architecture, which can output different classification results for a certain input, whereas the DL layer is formed by DL networks, which can continue learning from the outputs of the SL layer. DSL simulates a human thinking model that continuously learns from the existing learnings to improve learning efficiency. In our experiment, three shallow classification algorithms, i.e., C4.5, k-nearest neighbor, and naive Bayesian, are used to train base classifiers in the SL layer, whereas a deep belief network (DBN) is used to train the DL layer. The experiment results on three different datasets indicate that DSL outperforms other methods in terms of classification accuracy by using backpropagation neural network, bagging, AdaBoost, random forest, multilayer perceptron, and DBN.

入藏号: WOS:000633393000005

语言: English

文献类型: Article

作者关键词: Licenses; Classification algorithms; Feature extraction; Training; Recurrent neural networks; Spatial resolution; Remote sensing; Deep– shallow learning (DSL); deep learning (DL); ensemble learning (EL); image classification; remote sensing (RS)

地址: [Dou, Peng; Shen, Huanfeng; Li, Zhiwei; Guan, Xiaobin; Huang, Wenli] Wuhan Univ, Sch Resource & Environm Sci, Wuhan 430079, Peoples R China.

通讯作者地址: Li, ZW (通讯作者)Wuhan Univ, Sch Resource & Environm Sci, Wuhan 430079, Peoples R China.

电子邮件地址: dp_imgclassifier@163.com; shenhf@whu.edu.cn; lizw@whu.edu.cn; guanxb@whu.edu.cn; lwenli.huang@whu.edu.cn

影响因子:3.827


信息服务
学院网站教师登录 学院办公电话 学校信息门户登录

版权所有 © 武汉大学资源与环境科学学院
地址:湖北省武汉市珞喻路129号 邮编:430079 
电话:027-68778381,68778284,68778296 传真:027-68778893    邮箱:sres@whu.edu.cn