基于分形和神经网络理论的多尺度图象分割方法

杨绍国; 尹忠科; 罗炳伟

基于分形和神经网络理论的多尺度图象分割方法

计量
- 文章访问数: 2360
- HTML全文浏览量: 129
- PDF下载量: 388
- 被引次数: 0
出版历程
- 收稿日期: 1997-01-20
- 修回日期: 1998-01-15
- 刊出日期: 1998-11-19

MULTISCALE IMAGE SEGMENTATION USING FRACTAL AND NEURAL NETWORK

摘要

摘要: 特征空间聚类分割方法存在的关键问题是有效的特征参数提取和聚类方法的构造。针对这两个问题,本文采用小波变换的多尺度分析方法提取图象的多尺度分形维数作为分割特征参数,用Kohonen自组织特征映射实现特征空间聚类,获得了良好的分割效果。
- 分形; 小波变换; 神经网络; 图象分割
Abstract: Clustering algorithms in feature space are important methods in image segmentation. The choice of the effective feature parameters and the construction of the clustering method are key problems encountered with clustering algorithms. In this paper we chose multifractal dimensions as the segmentation feature parameters which are extracted from original image and wavelet-transformed image. SOM network is applied to cluster the segmentation feature parameters. The experiment shows that the performance of our algorithm is very good.

HTML全文

参考文献(1)

Ishimura N. Estimation of fractal dimension values in natural navigation images and its application to region segmentation. Joural of Japan Institute of Navigation, 1994, 90(3): 43-51.[2]Fortin C S. Fractal dimension in the analysis of medical images. IEEE Engineering in Medicine and Biology Xlagazine, 1992, 11(，)65-71.[3]Chang J. Image segmentation (IS) and local fractal analyses of MR images. Conference Record of the 1992 IEEE Nuclear Science Symposium and Medical Imaging Conference, Orlando, FL, USA: 1992, 2: 1268-1273.[4]Lefebvre F. A fractal approach to the segmentation of microcalcifications in digital mammograms[J].Medical Physics.1995, 22(4):381-390[5]Wong S H. Automatic segmentation of ultrasonic image. Proceedings TENCON93, 1993 IEEE Region 10 Conference on `Computer, Communication, Control and Power Engineering, Beijing, China: 1993, Vo1.2, 910-913.[6]Chan K L. Quantitative characterization of electron micrograph image using fractal feature[J].IEEE Trans. on Biomedical Engineering.1995, 42(10):1033-1037[7]Moghaddam B. Ractal dimension segmentation of synthetic aperture radar imagery. ISSPA 92, Third International Symposium on Signal Processing and its Application, Proceedings, Gold Coast, Auslralia: 1992, 455-458[8]朱光喜, 张平, 朱烟庭. 基于分形维数的图象分割研究.计算机科学, 1994, 21(1): 59-65.[9]罗立民,等.基于纹理分析的磁共振图象区域分割.自动化学报,1995, 21(4): 504-508.[10]Xue Dong-hui. The object detection based on multiscale fractal character vector. IEEE International Conference on Neural Networks and Signal Processing, 1995, 1451-1454.[11]Pentland A P. Fractal-based description of natural scences. IEEE Trans. on Pattern Analysis[12]and Machine Intelligence, 1984, PAMI-6(6): 661-674.[13]Peli T. Multiscale fractal theory and object characterization[J].J. Opt. Soc. Am. A.1990, 7(6):1101-1112[14]Dubuisson M P. Efficacy of fractal features in segmenting images of natural textures[J].Pattern Recognition Letters.1994, 15(4):419-431[15]Kasparis T. Texture description using fractal and energy features[J].Computers Electrical Engineering.1995, 21(1):21-32[16]Chaudhuri B B. Texture segmentation using fractal dimension. IEEE Trans. on Pattern Analysis and Machine Intelligence, 1995, PAMI-17(1): 72-77.[17]Mallat S G. A Theory for multiresolution signal decomposition: The wavelet representation. IEEE Trans. on Pattern Analysis and Machine Intelligence, 1989, PAMI-11(7): 674-693.[18]Arneodo A. Wavelet transform of multifractals[J].Physical Review Letters.1988, 61(20):2281-2284[19]Xuegong Zhang. Self-organizing map as a new method for clustering and data analysis. Proceed-[20]ings of 1993 International Joint Conference on Neural Networks, Nagoya, Japan: 1993, 2448-2451.[21]Witoon Suewatanakul. Comparison of artificial neural networks and traditional classifiers via the two-spiral problem. SPIE, 1992, 1721: 275-282.

施引文献

资源附件(0)

访问统计