层次信息量的线要素化简算法评价研究.pdfVIP

第 42 卷第 5 期 2013 年 10 月 测绘学报 Vo l. 42 , NO.5 Oc t. ,2013 Acta Geodaetica et Cartographica Sinica DENG Min , FAN Zide , LIU Huimin. Performance Evaluation of Line Simplification Algorithms Based on Hierarchical Information Con tent[J]. Acta Geodaetica et Cartographica Sinica , 2013 ,42( 5) : 767-773 , 781. (邓敏，樊子德，刘慧敏.层次信息量的线要素化简算法评价 研究[J].测绘学报， 2013 ， 42(5) :767-773 ,781.) 层次信息量的线要素化简算法评价研究 邓敏，樊子德，刘慧敏 中南大学地球科学与信息物理学院，湖南长沙 410083 Performance Evaluation of Li ne Simplificatio 门 Algorithms Based on Hierarchical Information Content DENG Min , FAN Zide , Ll U Huimin Department of Geo-informatics, Central South University , Changsha 410083 , China Abstract :Simplification has always been a commonly-used generalization operator. It is natural to further pay more attention to the distortions caused by generalization. The study takes line feature as an example and develops a new evaluation indicator (i. e. i 门formation conten t) at three levels (i. e. eleme川， neighborhood and holistic levels) from the view of inlormation transmission. The process 01 performance evaluation mainly involves the calculation and the compar巴on of the information content before and after line simplilication. For the lormer , the quantitative computational methods 01 inlormation content at the three levels are presented. For the latter , the diflerence 01 inlormation content belore and after line simplification is computed and utilized to measure the perlormance 01 line simplification algorithms. The difference of inlormation content at element level rellects the ability 01 line simplilication algorithms to select key points; the difference 01 information content at neighborhood level reflects the ability 01 line simplilication algorithms to mal 门tain bends; the diflerence 01 information co门tent at holistic level reflects the ability of line simplilication algorithm to maintain the trend 01 line leatures. Finally , a river 门etwork dataset is used to test the performance evaluations