改进利用蚁群规则挖掘算法进行遥感影像分类  

Remote Sensing Image Classification Based on Improved Ant-miner

作　　者：吴孔江[1] 曾永年[1] 靳文凭[1] 何丽丽[1] 李静[1] 

机构地区：[1]中南大学地球科学与信息物理学院空间信息技术与可持续发展研究中心,湖南长沙410083

出　　处：《测绘学报》2013年第1期59-66,共8页Acta Geodaetica et Cartographica Sinica

基　　金：国家自然科学基金(41171326;40771198);湖南省自然科学基金(08JJ6023)

摘　　要：基于ant-miner算法,提出改进蚁群规则挖掘算法。首先,从信息素浓度增加项、信息素挥发系数两方面,改进信息素浓度更新策略;其次,在算法求解中,引入变异算子,有效加快进化过程,缩短计算时间,获得较好的分类规则。以长沙市城区2006年TM影像为试验数据,在分类试验中对算法进行了验证。结果表明,相对于ant-miner和决策树方法而言,改进蚁群规则挖掘算法能挖掘出规则数目更少、形式更简单的分类规则,同时缩短计算时间,从而能够提高分类精度和效率。A new ant colony algorithm based on conventional ant-miner algorithm is proposed.Firstly,the conventional ant-miner algorithm is modified by using new pheromone concentration update item and pheromone evaporation coefficient.Then,the mutation operator is introduced in algorithm solve.The new ant colony algorithm accelerates effectively the evolutionary process and shorten the calculation time.In order to verify the new algorithm,the Landsat TM image of Changsha city is chosen as a case study area.The results indicate that the new algorithm obtains the simpler forms of classification rule and reduces the computation time.The remote sensing image classification using improved ant-miner algorithm is more accurate and efficient than conventional Ant-miner algorithm and decision tree.

关 键 词：蚁群规则挖掘 信息素更新 变异算子 遥感影像分类 

分 类 号：P208[天文地球—地图制图学与地理信息工程]