基于深度学习特征融合的遥感图像场景分类应用
作者单位:

中北大学

基金项目:

国家自然科学基金项目(面上项目,重点项目,重大项目)


Remote Sensing Image Scene Classification Application Based on Deep Learning Feature Fusion
Affiliation:

中北大学

Fund Project:

The National Natural Science Foundation of China (General Program, Key Program, Major Research Plan)

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献
  • | |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    针对传统手工特征方法无法有效提取整体图像深层信息,提出基于深度学习特征融合的场景分类新方法。首先利用灰度共生矩阵(GLCM)和局部二值模式(LBP)提取具有相关空间特性的纹理特征和局部纹理特征的浅层信息;其次通过基于AlexNet迁移学习网络提取图像的深层信息,在去除最后一层全连接层的同时加入一层256维的全连接层作为特征输出;将两种特征进行自适应融合;最终输入到网格搜索算法优化的支持向量机(GS-SVM)中对遥感图像进行场景分类识别。在公开数据集UC Merced的21类目标数据和RSSCN7的7类目标数据的实验结果表明,五次实验的平均准确率分别达94.77%和93.79%。实验结果与参考文献其它方法对分类精度进行对比,表明该方法有效提升遥感图像场景的分类精度。

    Abstract:

    In view of the traditional manual feature method, which can not effectively extract the overall image deep information, a new method of scene classification based on deep learning feature fusion is proposed. Firstly, the grayscale symbiotic matrix (GLCM) and local two-value pattern (LBP) are used to extract the shallow information of texture features and local texture features with relevant spatial characteristics, and secondly, the deep information of images is extracted by the AlexNet migration learning network, and a 256-dimensional full-connect layer is added as feature output while the last layer of full connection layer is removed, and the two features are adaptively integrated The remote sensing images are classified and identified in the support vector machine (GS-SVM) optimized by the grid search algorithm. The experimental results of the 21 types of target data of the public dataset UC Merced and the 7 types of target data of RSSCN7 show that the average accuracy rate of the five experiments was 94.77% and93.79%, respectively. The experimental results are compared with other methods in the references and the classification accuracy under the same data conditions, which shows that the proposed method can effectively improve the classification accuracy of remote sensing image scenes.

    参考文献
    [1] Zhao, Bei & Zhong, Yanfei & Xia, Gui-Song & Zhang, Liangpei. (2015). Dirichlet-derived multiple topic scene classification model fusing heterogeneous features for high resolution remote sensing imagery. IEEE Transactions on Geoscience and Remote Sensing. 54. 10.1109/TGRS.2015.496185.
    [2] Chen, Shizhi & Tian, Yingli. (2015). Pyramid of Spatial Relatons for Scene-Level Land Use Classification. Geoscience and Remote Sensing, IEEE Transactions on. 53. 10.1109/TGRS.2014.2351395.
    [3] D. Yu, H. Guo, Q. Xu, J. Lu, C. Zhao and Y. Lin, "Hierarchical Attention and Bilinear Fusion for Remote Sensing Image Scene Classification," in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 13, pp. 6372-6383, 2020, doi: 10.1109/JSTARS.2020.3030257.
    [4] 李孚煜,叶发茂.基于SIFT的遥感图像配准技术综述[J].国土资源遥感,2016,(02):14-20.(LI Fuyu,YE Famao. A Review of Remote Sensing Image Registration Technology Based on SIFT[J].Remote Sensing for Land and Resources,2016,(02):14-20.)
    [5] 许君一,徐富宝,张雅琼,彭代亮,聂忆黄,李晓红,范海生,张赫林,张强.基于灰度共生矩阵的未利用地疑似污染遥感识别[J].北京工业大学学报,2018,44(11):1423-1433.(XU Junyi,XU Fubao,ZHANG Yaqiong,PENG Dailiang,NIE Yihuang,LI Xiaohong,FAN Haisheng,ZHANG Helin,ZHANG Qiang. Remote sensing identification of suspected pollution in unused land based on gray scale symbiosis matrix[J].Journal of Beijing University of Technology,2018,44(11):1423-1433.)
    [6] 慕彩红,霍利利,刘逸,等.基于小波融合和PCA-核模糊聚类的遥感图像变化检测[J].电子学报,2015,(07):1375-1381.(Mu Caihong,Hollyley,Liu Yi,et al. Remote Sensing Image Change Detection Based on Wavelet Fusion and PCA-Nuclear Fuzzy Clustering[J].Acta Electronica Sinica,2015,(07):1375-1381.)
    [7] 张庆春,佟国峰,李勇,等.基于多特征融合和软投票的遥感图像河流检测[J].光学学报,2018,(06):320-326.(ZHANG Qingchun,TONG Guofeng,LI Yong,et al. River Detection of Remote Sensing Images Based on Multi-Feature Fusion and Soft Voting[J].Acta Optica Sinica,2018,(06):320-326.)
    [8] 王钰,何红艳,谭伟,齐文雯.一种多特征融合的高分辨率遥感图像道路提取算法[J].遥感信息,2019,34(01):111-116.(WANG Yu,HE Hongyan,TAN Wei,QI Wenwen. A Road Extraction Algorithm for High-Resolution Remote Sensing Images with Multi-Feature Fusion[J].Remote Sensing Information,2019,34(01):111-116.)
    [9] Han, X.; Zhong, Y.; Cao, L.; Zhang, L. Pre-Trained AlexNet Architecture with Pyramid Pooling and Supervision for High Spatial Resolution Remote Sensing Image Scene Classification. Remote Sens. 2017, 9, 848.
    [10] E. Li, J. Xia, P . Du, C. Lin, and A. Samat, “Integrating multilayer features of convolutional neural networks for remote sensing scene classification,” IEEE Trans. Geosci. Remote Sens., vol. 55, no. 10, pp. 5653–5665,Oct. 2017.
    [11] X. Lu, H. Sun, and X. Zheng, “A feature aggregation convolutional neural network for remote sensing scene classification,” IEEE Trans. Geosci.Remote Sens., vol. 57, no. 10, pp. 7894–7906, Oct. 2019.
    [12] 钱晓亮,李佳,程塨,等.特征提取策略对高分辨率遥感图像场景分类性能影响的评估[J].遥感学报,2018,(05):758-776.(Qian Xiaoliang,Li Jia,Cheng Yuan,et al. Evaluation of the effect of feature extraction strategy on scene classification performance of high-resolution remote sensing images[J].Journal of Remote Sensing,2018,(05):758-776.)
    [13] 聂光涛,黄华.光学遥感图像目标检测算法综述[J].自动化学报,2021,47(08)(NIE Guangtao,HUANG Hua. A Review of Optical Remote Sensing Image Object Detection Algorithms[J].Chinese Journal of Automation,2021,47(08).)
    [14] 杨帆.数字图像处理与分析.北京:航空航天大学出版 社, 2007. 277-281.(YANG Fan. Digital Image Processing and Analysis. Beijing:University of Aeronautics and Astronautics Press, 2007. 277-281.)
    [15] 余东行,张保明,赵传,等.联合卷积神经网络与集成学习的遥感影像场景分类[J].遥感学报,2020,(06):717-727(Yu Dongxing,Zhang Baoming,Zhao Chuan,et al. Remote Sensing Image Scene Classification based on Convolutional Neural Network and Ensemble Learning[J].Journal of Remote Sensing,2020,(06):717-727)
    [16] Krizhevsky, A. , Sutskever, I. , & Hinton, G. . (2012). ImageNet Classification with Deep Convolutional Neural Networks. NIPS (Vol.25). Curran Associates Inc.
    [17] 张成,侯宇超,焦宇倩,白艳萍,李建军.基于三通道分离特征融合与支持向量机的混凝土图像分类研究[J].图学学报,2021,42(06):917-923.(ZHANG Cheng,HOU Yuchao,JIAO Yuqian,BAI Yanping,LI Jianjun. Research on Concrete Image Classification Based on Three-Channel Separation Feature Fusion and Support Vector Machine[J].Journal of Graphics,2021,42(06):917-923.)
    [18] Zhang, Fan & Du, Bo & Zhang, Liangpei. (2015). Saliency-Guided Unsupervised Feature Learning for Scene Classification. Geoscience and Remote Sensing, IEEE Transactions on. 53. 10.1109/TGRS.2014.2357078.
    [19] 张艳月,张宝华,赵云飞,等.基于双通道深度密集特征融合的遥感影像分类[J].激光技术,2021,(01):73-79.(ZHANG Yanyue,ZHANG Baohua,ZHAO Yunfei,et al. Classification of remote sensing images based on two-channel deep-density feature fusion[J].Laser Technology,2021,(01):73-79.)
    [20] Negrel, Romain & Picard, David & Gosselin, Philippe-Henri. (2014). Evaluation of Second-order Visual Features for Land-Use Classification. Proceedings - International Workshop on Content-Based Multimedia Indexing. 1-5.10.1109/CBMI.2014.6849835.
    [21] Zhao, Bei & Zhong, Yanfei & Zhang, Liangpei & Huang, Bo. (2016). The Fisher Kernel Coding Framework for High Spatial Resolution Scene Classification. Remote Sensing. 8. 157.10.3390/rs8020157.
    [22] 许夙晖,慕晓冬,赵鹏,等.利用多尺度特征与深度网络对遥感影像进行场景分类[J].测绘学报,2016,(07):834-840.(Xu Suhui,Mu Xiaodong,Zhao Peng,et al. Scene classification of remote sensing images using multi-scale features and depth networks[J].Journal of Geomatics,2016,(07):834-840)
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

王李祺,张成,侯宇超,谭秀辉,程蓉,高翔,白艳萍.基于深度学习特征融合的遥感图像场景分类应用[J].南京信息工程大学学报,,():

复制
分享
文章指标
  • 点击次数:147
  • 下载次数: 0
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 收稿日期:2022-03-22
  • 最后修改日期:2022-05-05
  • 录用日期:2022-05-09

地址:江苏省南京市宁六路219号    邮编:210044

联系电话:025-58731025    E-mail:nxdxb@nuist.edu.cn

南京信息工程大学学报 ® 2025 版权所有  技术支持:北京勤云科技发展有限公司