| 118 | 0 | 12 |
| 下载次数 | 被引频次 | 阅读次数 |
遥感影像数据挖掘是一个有着广阔应用前景的研究领域。对图像检索、图像分类、图像聚类、空间关联规则挖掘和图像变化检测等数据挖掘应用而言,相似性度量是基础和前提。采用了图像空间划分的策略,在此基础上计算颜色、纹理和形状等3方面的低层视觉特征来描述图像,采用多维特征空间的网格划分来降低数据维数并建立了影像的相似性度量。实验结果表明,该方法对影像具有一定的几何和光照不变性。
Abstract：Remote sensing image data mining is a research field with wide application prospect.For the data mining application such as image searching,image classing,image clustering,space associated rule mining,image change detecting,etc.,the similarity measurement is the necessary condition and foundation.The paper introduces the strategy of image space partition,gives a image description by using the low-level vision feature of color,texture and shape etc,reduces the data dimension with the mesh partition of multidimensional feature space,and establishes the similarity measurement of image.The experimental results show that this method has a certain extent geometry and optics invariability.
[1]李德仁,王树良,史文中,等.论空间数据挖掘与知识发现[J].武汉大学学报(信息科学版),2001,26(6):491-499.
[2]GESUD V,STAROVOITOVV.Distance-based Functions for Image Comparison[J].Pattern Recognition Letters,1999(20):207-214.
[3]王润生.图像理解[M].长沙:国防科技大学出版社,1995.
[4]TAMURA H,MORI S,YAMAWAKI T.Texture Features Correspondingto Visual Perception[J].IEEE Trans on SMC,1978,8(6):460-473.
[5]JAINAK,FARROKNIAF.Unsupervised Texture Segmentation Using Gabor Filters[J].Pattern Recognition,1991,24(12):1167-1186.
[6]LIENHART R.Comparison of Automatic Shot Boundary Detection Algorithm[C].USA:In Proc.of SPIE Storage and Retrieval for Image and Video Databases VII,1999:291-301.
[7]贺玲,吴玲达.多媒体数据挖掘中数据间的相似性度量研究[J].国防科技大学学报,2006,28(1):77-80.
[8]COLEMAN G B,ANDREWS H C.Image Segmentation by Clustering[J].Proceeding of IEEE,1979,67(5):773-785.
基本信息:
DOI：
中图分类号:TP751
引用信息:
[1]陈晓飞,段洣毅,张佳骥.一种遥感影像挖掘的相似性度量方法[J].无线电工程,2008,38(12):29-31.
基金信息: