• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学

• 人工智能与数据挖掘 • 上一篇    下一篇

基于特征整合的卷积神经网络草地分类算法

张猛,钱育蓉,杜娇,范迎迎   

  1. (新疆大学软件学院,新疆 乌鲁木齐 830046)
  • 收稿日期:2017-12-25 修回日期:2018-04-11 出版日期:2019-07-25 发布日期:2019-07-25
  • 基金资助:

    国家自然科学基金(61562086,61363083)

A grassland classification algorithm using convolutional
neural network based on feature integration

ZHANG Meng,QIAN Yurong,DU Jiao,FAN Yingying   

  1. (Software College,Xinjiang University,Urumqi 830046,China)
     
  • Received:2017-12-25 Revised:2018-04-11 Online:2019-07-25 Published:2019-07-25

摘要:

为提高遥感影像草地分类的精度,分析了卷积神经网络中提取图像特征的特点,提出了一种基于特征整合深度神经网络的遥感影像特征提取算法。首先,将遥感影像数据进行PCA白化处理,降低数据之间的相关性,加快神经网络学习的速率;其次,将从卷积神经网络中提取到的浅层特征和深层特征进行双线性整合,使得整合后的新特征更加完善和优化;最后,对遥感数据进行训练,由于新特征中有效信息的增加,使得特征表达能力得到提高,达到提高草地分类准确率的目的。实验结果表明:该算法能够有效地提高草地分类的准确率,分类精度达到94.65%,相较于卷积神经网络、BP神经网络和基于SVM的分类算法分别提高了4.3%、10.39%和15.33%。

 

关键词: 遥感影像, 草地分类, 卷积神经网络, 特征整合, PCA白化

Abstract:

In order to improve the precision of grassland classification from remote sensing images, we analyze the characteristics of image features extracted from convolutional neural networks (CNNs), and propose a remote sensing image feature extraction method based on feature-integrated depth neural networks. Firstly, PCA whitening is performed on the remote sensing image to reduce the correlation between data and accelerate the learning rate of neural networks. Secondly, both low-level features and high-level features are bilinearly integrated to enhance and optimize the integrated features. Finally, the remote sensing data is trained. As the introduction of effective information in new features, both feature expression ability and the grassland classification accuracy are improved. Experimental results show that the proposed algorithm can effectively improve the accuracy of grassland classification. The classification accuracy reaches up to 94.65%. Compared with the traditional convolutional neural network, BP neural network and SVM algorithm, our accuracy is increased by 4.3%, 10.39% and 15.33% respectively.
 
 

Key words: remote sensing image, grassland classification, convolutional neural network, integrated feature, PCA whitening