基于特征选择的土地整理地物识别规则的动态提取  

The Dynamic Extraction of Classification Rules for Land Consolidation through Object-oriented Features Space Optimization

作　　者：吴健生[1,2] 林倩[1] 李卫锋[3] 刘建政[3] 彭建[2] 

机构地区：[1]北京大学深圳研究生院城市人居环境科学与技术重点实验室,广东深圳518055 [2]北京大学城市与环境学院地表过程与模拟教育部重点实验室,北京100871 [3]香港大学城市规划与设计系

出　　处：《遥感技术与应用》2013年第5期799-806,共8页Remote Sensing Technology and Application

基　　金：国家自然科学基金项目(41271101)资助

摘　　要：高分辨率遥感影像是精确提取土地整理区地表信息的重要数据来源,提出了一种基于面向对象的规则动态提取方法,借鉴生物学中的遗传定理和人工免疫系统理论,使计算机依据样本信息和影像特征自动进行知识挖掘,提取分类规则以供用户修改与分类。最后结合模糊分类得到的分类结果显示:总体精度从传统方法的40%提高到基于遗传算法的86%与基于人工免疫算法的90%,Kappa系数也由传统方法的0.3提高到基于遗传算法的0.82与基于人工免疫算法的0.89。结果表明:该方法不仅提高了便捷性与通用性,改变了以往规则提取需要用户大量的试验和先验知识的局面,而且试验结果也表明对于分类精度有显著的提高,对于在土地整理工程中利用高分辨率遥感影像进行地物识别与监测有重要的意义。The high resolution remote sensing image is an important data sources for the of land consolidation area surface information. In this paper,a new object-based method, netic algorithm and artificial immune algorithm,is used to extract classification rules b teristics of the sample image. After fuzzy classification,the results show that overall ac from 40% by traditional method to 86% corresponding to the genetic algorithm and 90 the artificial immune algorithm,and the Kappa coefficient is increasing from 0.3 by tra 0.82 corresponding to the genetic algorithm and 0.89 corresponding to the artificial accurate extraction combining with geased on the charac- curacy is increasing 90% corresponding to ditional methods to algorithm. All in all, not only this method can improve the convenience and versatility, changing the previous situation that the rule extraction requires users a large amount of priori knowledge and testing,but also the test re sults show the significant improvement in classification accuracy. Therefore,it has an important significance for land consolidation,especially using the high-remote sensing images for feature identifying and ring.

关 键 词：土地整理 特征选择 高分辨率遥感影像 面向对象 遗传算法 人工免疫算法 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]