| 基于改进支持向量机方法的影像分类和目标识别研究 |
| |
| 引用本文： | 贺卫中,周维娜,束平. 基于改进支持向量机方法的影像分类和目标识别研究[J]. 遥感信息, 2010, 0(6): 6-8,13. DOI: 10.3969/j.issn.1000-3177.2010.06.002 |
| |
| 作者姓名： | 贺卫中 周维娜 束平 |
| |
| 作者单位： | 常州市土地勘测中心,江苏省基础地理信息中心常州分中心,常州213001 |
| |
| 基金项目： | 江苏省测绘局2009年度测绘科研基金资助项目 |
| |
| 摘 要： | 利用卫星遥感影像进行土地利用变化监测的关键技术是影像分类与目标识别。本文提出了支持向量机的改进算法,基于小波核函数构建了小波模糊支持向量机。通过项目"集成卫星遥感与地形地籍数据进行土地利用变化检测"的研究和实验,力求在创新处理算法上取得突破,提高重点目标识别的准确性、可靠性。
|
| 关 键 词： | 小波核函数 模糊支持向量机 目标识别 |
Research on Classification and Target Recognition of Remote Sensing Image Based on Improved Support Vector Machine |
| |
| HE Wei-zhong,ZHOU Wei-na,SU Ping. Research on Classification and Target Recognition of Remote Sensing Image Based on Improved Support Vector Machine[J]. Remote Sensing Information, 2010, 0(6): 6-8,13. DOI: 10.3969/j.issn.1000-3177.2010.06.002 |
| |
| Authors: | HE Wei-zhong ZHOU Wei-na SU Ping |
| |
| Affiliation: | HE Wei-zhong,ZHOU Wei-na,SU Ping(Land Resource Surveying Center of Changzhou,Jiangsu Piovincial Fundamental Geomatics Center of Changzhou,Changzhou 213001) |
| |
| Abstract: | In order to detect and monitor the change of land use,a novel method based on satellite remote sensing image processing and object recognition is proposed.The standard support vector machine is improved by introducing wavelet kernel function and fuzzy theory,producing wavelet fuzzy support vector machine（WFSVM）.The remote sensing imagery is classified by WFSVM.Furthermore,water,building and roadway are extracted and three theme layers are created via the object extracted.This method is applied in a practical project and the result demonstrates that the method is robust and with high accuracy. |
| |
| Keywords: | wavelet kernel function fuzzy support vector machine object recognition |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
