摘要
作为K-means算法的优化算法,Mini Batch K-means算法在遥感影像分类中的应用较少.分别利用Mini Batch K-means算法与K-means算法对10个不同幅度的EVI遥感影像数据进行分类.对比两种分类算法的精度和时间复杂度发现,相比于K-means算法,Mini Batch K-means虽然损失了小部分的精度,但却极大提高了分类效率,更适用于大数据量的遥感影像分类.
As an optimization algorithm for K-means algorithm,Mini Batch K-means algorithm was less used in remote sensing image classification. The Mini Batch K-means and K-means algorithms were used to investigate the land cover classification based on 10 different extent EVI data. The classification accuracy and time complexity were compared and the results showed that the Mini Batch K-means algorithm greatly improved the efficiency of classification though it lost a small part of accuracy. This algorithm was suitable for the classification of remote sensing data.
出处
《鲁东大学学报（自然科学版）》
2017年第4期359-363,共5页
Journal of Ludong University:Natural Science Edition
基金
中国博士后基金(2015M572061)
全国统计科学研究项目(2016LY76)
山东省自然科学基金面上项目(ZR2015FM014)
