吉林大学学报(地球科学版) ›› 2018, Vol. 48 ›› Issue (2): 373-378.doi: 10.13278/j.cnki.jjuese.20170271

• 地球物理数据处理与解释技术 • 上一篇    下一篇

基于极限学习机的GF-2影像分类

王明常1,2,3, 张馨月1, 张旭晴1, 王凤艳1, 牛雪峰1, 王红2   

  1. 1. 吉林大学地球探测科学与技术学院, 长春 130026;
    2. 湖北大学资源环境学院, 武汉 430062;
    3. 国土资源部城市土地资源监测与仿真重点实验室, 广东 深圳 518000
  • 收稿日期:2017-09-07 出版日期:2018-03-26 发布日期:2018-03-26
  • 通讯作者: 王红(1975-),女,副教授,博士,主要从事遥感与地理信息系统方面的教学研究,E-mail:j-wanghong@163.com E-mail:j-wanghong@163.com
  • 作者简介:王明常(1975-),男,教授,博士,主要从事遥感与地理信息系统方面的教学研究,E-mail:wangmc@jlu.edu.cn
  • 基金资助:
    国家自然科学基金项目(41430322);国土资源部城市土地资源监测与仿真重点实验室开放基金项目(KF-2016-02-008);区域开发与环境响应湖北省重点实验室开放研究基金项目(2015(B)003)

GF-2 Image Classification Based on Extreme Learning Machine

Wang Mingchang1,2,3, Zhang Xinyue1, Zhang Xuqing1, Wang Fengyan1, Niu Xuefeng1, Wang Hong2   

  1. 1. College of GeoExploration Science and Technology, Jilin University, Changchun 130026, China;
    2. Faculty of Resources and Environment, Hubei University, Wuhan 430062, China;
    3. Key Laboratory of Urban Land Resources Monitoring and Simulation, Ministry of Land and Resources, Shenzhen 518000, Guangdong, China
  • Received:2017-09-07 Online:2018-03-26 Published:2018-03-26
  • Supported by:
    Supported by National Natural Science Foundation of China(41430322), Open Fund of Key Laboratory of Urban Land Resources Monitoring and Simulation, Ministry of Land and Resources (KF-2016-02-008) and Open Research Fund Program of Hubei Province Key Laboratory of Regional Development and Environmental Response(2015(B)003)

摘要: 遥感图像分类是提取图像有效信息过程中重要的一部分,为了探寻最优的分类方法,许多机器学习算法逐步应用于遥感分类中。极限学习机(extreme learning machine,ELM)以其高效、快速和良好的泛化性能在模式识别领域得到广泛应用。本文采用训练速度快、运算量小的极限学习机算法与支持向量机(support vector machines,SVM)算法和最大似然法进行分类对比,对高分辨率遥感图像进行分类,分析极限学习机算法对于遥感图像分类的准确度等性能。选取吉林省长春市部分区域的GF-2遥感数据,将融合后的影像设置为原始数据,利用3种方法进行分类。研究结果表明,极限学习机算法分类图像总体分类精度达到85%以上,kappa系数达到0.718,与其他分类方法相比分类准确度较高,且极限学习机运行时间比支持向量机运行时间约短2 480 s,约为支持向量机运行时间的1/8,因此具有良好的性能和实用价值。

关键词: 极限学习机, 遥感图像分类, GF-2影像, 监督分类, 支持向量机

Abstract: The classification of remote sensing image is an important part of extracting effective information of images. In order to explore the optimal classification methods, many machine learning algorithms are gradually applied to the classification of remote sensing images. Because of the high efficiency, speediness, and good performance generalization, extreme learning machines (ELM) have been widely used in pattern recognition. This paper aims to classify high-resolution remote sensing images, and analyze the performance of extreme learning machine algorithms for the accuracy of classification of remote sensing images, and compare ELM algorithm with support vector machines (SVM) algorithm and Maximum likelihood method. The GF-2 data from some area in Changchun City were selected to test the accuracy of all the three methods for classification with the fused image as the original data. The results show that the overall accuracy of ELM algorithm is more than 85%, and the kappa coefficient is 0.718. Compared with the other two classification methods, the ELM classification accuracy is the best, and its running time is faster than the support vector machine by 2 480 s, 1/8 of the support vector machine's. An even better performance can be obtained in a shorter training time. ELM is valuable for classification of remote sensing images.

Key words: extreme learning machine, remote sensing image classification, GF-2 image, supervised classification, support vector machines

中图分类号: 

  • P237
[1] 陈圣波,刘彦丽,杨倩,等.植被覆盖区卫星高光谱遥感岩性分类[J].吉林大学学报(地球科学版),2012,42(6):1959-1965. Chen Shengbo,Liu Yanli,Yang Qian,et al. Lithologic Classification from Hyperspectral Data in Dense Vegetation Cover Area[J].Journal of Jilin University (Earth Science Edition),2012,42(6):1959-1965.
[2] Nitish S,Geoffrey H,Alex K,et al. Dropout:A Simple Way to Prevent Neural Networks from Overfitting[J].Journal of Machine Learning Research,2014,15(6):1929-1958.
[3] Huang G B,Zhu Q Y,Siew C K. Extreme Learning Machine:A New Learning Scheme of Feedforward Neural Networks[J]. Proc Int Joint Conf Neural Netw,2004,2(2):985-990.
[4] Zong W W,Huang G B,Chen Y Q. Weighted Extreme Learning Machine for Imbalance Learning[J]. Neurocomputing,2013,101:229-242.
[5] Huang G B,Ding X J,Zhou H M. Optimization Me-thod Based Extreme Learning Machine for Classification[J]. Neurocomputing,2010,74:155-163.
[6] Zong W W,Huang G B. Face Recognition Based on Extreme Learning Machine[J].Neurocomputing,2011,74(16):2541-2551.
[7] Huang G B,Li M B,Chen L,et al. Incremental Extreme Learning Machine with Fully Complex Hidden Nodes[J]. Neurocomputing,2008,71(4):576-583.
[8] 王士同,钟富礼.最小学习机[J].江南大学学报(自然科学版),2010,9(5):505-510. Wang Shitong,Zhong Fuli. On Least Learning Machine[J]. Journal of Jiangnan University (Natural Science Edition),2010,9(5):505-510.
[9] 邓万宇,郑庆华,陈琳,等.神经网络极速学习方法研究[J].计算机学报,2010,33(2):280-287. Deng Wanyu,Zheng Qinghua,Chen Lin,et al. Research on Extreme Learning of Neural Networks[J].Chinese Journal of Computer,2010,33(2):280-287.
[10] 王杰,毕浩洋.一种基于粒子群优化的极限学习机[J].郑州大学学报(理学版),2013,45(1):100-104. Wang Jie,Bi Haoyang. A New Extreme Learning Machine Optimized by PSO[J]. Journal of Zhengzhou University (Natural Science Edition),2013,45(1):100-104.
[11] 陈建原,何建农.基于在线连续极限学习机的图像分类改进算法[J].福州大学学报(自然科学版),2015,42(2):176-181. Chen Jianyuan,He Jiannong. A Modified Algorithm for Image Classification Based on Online Sequential Extreme Learning Machine[J].Journal of Fuzhou University (Natural Science Edition),2015,42(2):176-181.
[12] 王礼云,毛文涛,胡武鹏.极限学习机多目标模型选择研究[J].计算机仿真,2014,31(8):387-391. Wang Liyun,Mao Wentao,Hu Wupeng. Multi-Objective Model Selection for Extreme Learning Machine[J]. Computer Simulation,2014,31(8):387-391.
[13] 吴军,王士同,赵鑫.正负模糊规则系统、极限学习机与图像分类[J].中国图象图形学报,2011,16(8):1408-1417. Wu Jun,Wang Shitong,Zhao Xin. Positive and Negative Fuzzy Rule System,Extreme Learning Machine and Image Classification[J].Journal of Image and Graphics,2011,16(8):1408-1417.
[14] 崔利凯,孙建孟,闫伟超,等.基于多分辨率图像融合的多尺度多组分数字岩心构建[J].吉林大学学报(地球科学版),2017,47(6):1904-1912. Cui Likai,Sun Jianmeng,Yan Weichao,et al. Construction of Multi-Scale and Component Digital Cores Based on Fusion of Different Resolution Core Images[J].Journal of Jilin University (Earth Science Edition),2017,47(6):1904-1912.
[1] 卢文喜, 郭家园, 董海彪, 张宇, 林琳. 改进的支持向量机方法在矿山地质环境质量评价中的应用[J]. 吉林大学学报(地球科学版), 2016, 46(5): 1511-1519.
[2] 王常明, 田书文, 王翊虹, 阮云凯, 丁桂伶. 泥石流危险性评价:模糊c均值聚类-支持向量机法[J]. 吉林大学学报(地球科学版), 2016, 46(4): 1168-1175.
[3] 秦喜文, 刘媛媛, 王新民, 董小刚, 张瑜, 周红梅. 基于整体经验模态分解和支持向量回归的北京市PM2.5预测[J]. 吉林大学学报(地球科学版), 2016, 46(2): 563-568.
[4] 杨雪峰, 王雪梅, 毛东雷. 塔里木河下游土地利用覆被MISR多角度遥感制图[J]. 吉林大学学报(地球科学版), 2016, 46(2): 617-626.
[5] 牟丹, 王祝文, 黄玉龙, 许石, 周大鹏. 基于最小二乘支持向量机测井识别火山岩类型:以辽河盆地中基性火山岩为例[J]. 吉林大学学报(地球科学版), 2015, 45(2): 639-648.
[6] 陈永良,李学斌. 核概率距离聚类方法及应用[J]. 吉林大学学报(地球科学版), 2013, 43(1): 312-318.
[7] 杨佳佳, 姜琦刚, 赵静, 徐言, 孟翔冲. 基于改进的SVM技术和高光谱遥感的标准矿物定量计算[J]. J4, 2012, 42(3): 864-871.
[8] 牛瑞卿, 彭令, 叶润青, 武雪玲. 基于粗糙集的支持向量机滑坡易发性评价[J]. J4, 2012, 42(2): 430-439.
[9] 许长福, 李雄炎, 谭锋奇, 于红岩, 李洪奇. 任务驱动数据挖掘方法的提出及在低阻油层识别中的应用[J]. J4, 2012, 42(1): 39-46.
[10] 王利花, 周云轩, 田波. 基于TM和ETM+影像数据的东沙环礁珊瑚礁监测[J]. J4, 2011, 41(5): 1630-1637.
[11] 佴磊, 彭文, 袁明哲, 周能娟. 基于经验模态分解和加权最小二乘支持向量机的采空区地面塌陷预测[J]. J4, 2011, 41(3): 799-804.
[12] 李秀珍, 孔纪名, 王成华. 多分类支持向量机在滑坡稳定性判识中的应用[J]. J4, 2010, 40(3): 631-637.
[13] 徐红敏,杨天行. 基于支持向量机分类算法的湖泊水质评价研究[J]. J4, 2006, 36(04): 570-573.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!