| PCA和布谷鸟算法优化SVM的遥感矿化蚀变信息提取 |
| |
| 引用本文： | 吴一全,盛东慧,周杨.PCA和布谷鸟算法优化SVM的遥感矿化蚀变信息提取[J].遥感学报,2018,22(5):810-821. |
| |
| 作者姓名： | 吴一全 盛东慧 周杨 |
| |
| 作者单位： | 南京航空航天大学电子信息工程学院;中国地质科学院矿产资源研究所国土资源部成矿作用与资源评价重点实验室;国土资源部地质信息技术重点实验室;成都理工大学国土资源部地学空间信息技术重点实验室;兰州大学甘肃省西部矿产资源重点实验室;东华理工大学江西省数字国土重点实验室 |
| |
| 基金项目： | 国家自然科学基金（编号：61573183）；中国地质科学院矿产资源研究所国土资源部成矿作用与资源评价重点实验室开放基金项目（编号：ZS1406）；国土资源部地质信息技术重点实验室开放基金项目（编号：217）；成都理工大学国土资源部地学空间信息技术重点实验室开放基金项目（编号：KLGSIT2015-05）；兰州大学甘肃省西部矿产资源重点实验室开放基金项目（编号：WCRMGS-2014-05）；东华理工大学江西省数字国土重点实验室开放基金项目（编号：DLLJ201412） |
| |
| 摘 要： | 为了进一步提高遥感矿化蚀变信息提取的精度,本文提出了一种基于主成分分析PCA (Principal Component Analysis)和布谷鸟算法优化支持向量机SVM (Support Vector Machine)的遥感矿化蚀变信息提取方法。首先,通过波段比值法增强研究区遥感图像中的矿化蚀变信息,并获得比值图像;然后,对比值图像进行主成分分析,进而提取训练样本;接着,利用SVM对训练样本进行训练,同时采用布谷鸟算法求取SVM的最优核参数及惩罚因子,构造最优SVM模型;最后,运用最优SVM模型完成矿化蚀变信息提取。选择青海省五龙沟地区为研究区,提取羟基及铁染蚀变信息。实验结果表明,与主成分分析法、基于光谱角法和SVM的方法、基于粒子群和SVM的方法及基于波段比值、PCA和粒子群优化SVM的方法等4种方法相比,本文方法获得的遥感矿化蚀变信息和已知矿点的吻合度最高,提取效果最好。
|
| 关 键 词： | 遥感 矿化蚀变信息提取 主成分分析(PCA) 支持向量机(SVM) 布谷鸟算法 波段比值法 |
| 收稿时间： | 2017/3/21 0:00:00 |
Remote sensing mineralization alteration information extraction based on PCA and SVM optimized by cuckoo algorithm |
| |
| WU Yiquan,SHENG Donghui and ZHOU Yang.Remote sensing mineralization alteration information extraction based on PCA and SVM optimized by cuckoo algorithm[J].Journal of Remote Sensing,2018,22(5):810-821. |
| |
| Authors: | WU Yiquan SHENG Donghui and ZHOU Yang |
| |
| Institution: | College of Electronic and Information Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China;Key Laboratory of Metallogeny and Mineral Assessment, Ministry of Land and Resources, Institute of Mineral Resources, Chinese Academy of Geological Sciences, Beijing 100037, China;Key Laboratory of Geological Information Technology, Ministry of Land and Resources, Beijing 100037, China;Key Laboratory of Geo-Spatial Information Technology, Ministry of Land and Resources, Chengdu University of Technology, Chengdu 610059, China;Key Laboratory of Mineral Resources in Western China(Gansu Province), Lanzhou University, Lanzhou 730000, China;Digital Land Key Laboratory of Jiangxi Province, East China Institute of Technology, Nanchang 330013, China,College of Electronic and Information Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China and College of Electronic and Information Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China |
| |
| Abstract: | With the rapid development of the economy, the demand for mineral resources is growing, and the contradiction between supply and demand is increasing. The shortage of mineral resources has become one of the important factors that restrict national economic development. Therefore, research on how to efficiently and accurately explore mineral resources is a critical endeavor. Remote sensing mineralization alteration information extraction is an important application of remote sensing technology in geological exploration, which is of utmost significance to mineral exploration and evaluation. Owing to the influence of vegetation, cloud, and snow, alteration information from remote sensing mineralization is often superimposed with the complex geological background and exists only in the form of a weak signal in the background of the remote sensing image. Research on effective remote sensing mineralization alteration information extraction methods can provide the basis for the study of regional metallogenic prognosis and speed up the evaluation of mineral resources exploration, which helps promote the healthy and stable development of the local mining economy.To improve the accuracy of remote sensing mineralization alteration information extraction method, a remote sensing mineralization alteration information extraction method based on Principal Component Analysis (PCA) and Support Vector Machine (SVM) optimized by cuckoo algorithm is proposed in this study. First, the mineralization alteration information in the remote sensing image of the study area is enhanced by band ratio method, and the ratio images are obtained. Then PCA is applied to the ratio images of the study area. The hydroxyl principal components and iron staining principal components are selected, after which the training samples are extracted. Subsequently, the training samples are trained by SVM, while cuckoo algorithm is used to find the optimal kernel parameter and penalty factor of SVM. Thus, the optimal SVM model is determined. Finally, the optimal SVM model is used to accomplish the extraction of remote sensing mineralization alteration information in the study area.Wulonggou area of Qinghai Province, which is rich in mineral resources, is selected as the study area where the hydroxyl alteration information and iron alteration information are extracted. A detailed comparison among the proposed method and four methods proposed recently, namely, the PCA method, the method based on spectral angle mapper and SVM, the method based on particle swarm optimization and SVM, and the method based on band ratio, PCA, and SVM optimized by particle swarm optimization in terms of extraction effect and matching rate, is given in this paper. Experimental results show that by using the proposed method, the extracted information can comprehensively reflect the remote sensing mineralization alteration information of the study area. Moreover, the matching degree of hydroxyl alteration information and iron alteration information are 86.5% and 69.2%, respectively. Meanwhile, compared with the four methods, the proposed method can obtain the highest matching degree with the best extraction effect.The proposed remote sensing mineralization alteration information extraction method based on PCA and SVM optimized by cuckoo algorithm is an effective method that provides a new idea for mineral exploration and metallogenic prediction. |
| |
| Keywords: | remote sensing mineralization alteration information extraction principal component analysis support vector machine cuckoo algorithm band ratio method |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《遥感学报》浏览原始摘要信息 |
| 点击此处可从《遥感学报》下载免费的PDF全文 |
|
