基于多级小波分解的两类纹理图像分割技术

作者：夏敏磊1，郭斌2 日期：2007-10-24/span> 浏览：3888 查看PDF文档

基于多级小波分解的两类纹理图像分割技术

夏敏磊1，郭斌2
(1.浙江机电职业技术学院 电子信息工程系，浙江 杭州 310053;2.浙江省机电设计研究院有限公司，浙江 杭州 310009)

摘要：通过对纹理图像传统分割方法的分析，研究了农作物图像的边缘自动分割问题，提出了采用多级小波分解方法的两类纹理图像分割技术。解决了传统的基于Laws能量特征方法存在的分割边缘粗糙以及时间效率低等缺点。仿真结果表明，应用基于多级小波分解的纹理分割方法，能快速得到准确的边缘，为获取大规模农作物的机械导航路径提供了一种新的思路。
关键词：多级小波分解；纹理分割；机械导航
中图分类号：TP391.41文献标识码：B文章编号：1001-4551(2007)09-0115-02

Image segmentation for two types of texture based on
multilevel wavelet decomposition
XIA Minlei1, GUO Bin2
(1. Department of Electronic and Information Engineering, Zhejiang Institute of Mechanical & Electrical Engineering,
Hangzhou 310053, China; 2. Zhejiang Institute of Mechanical & Electrical Engineering Co. Ltd., Hangzhou 310009, China)
Abstract: Image segmentation technology for two types of texture based on multilevel wavelet decomposition method was proposed, which had solved the rough edge and split time low efficiency shortcomings based on the traditional methods, such as Laws energy characteristic, based on analyzing the traditional image texture segmentation method. Experimental results show that the method can rapidly obtain accurate edge based on multilevel wavelet decomposition and provide access to a new thought of mechanical navigation routines for a largescale crop.
Key words: multilevel wavelet decomposition; texture segmentation; machinery guidance
参考文献(Reference)：
［1］LIU Y, ZHOU X. Automatic Texture Segmentation for Texturebased Image Retrieval［C］. IEEE Proceeding of the 10th International Multimedia Modeling (MMM 04), 2004: 285-290.
［2］ZHANG Yuanlin, ZHENG Nanning, QUAN Wei. Unsupervised texture segmentation based on dynamic pyramid label random field modeling［J］. MiniMicro Systems, 2005,26(3):322-325.
［3］KUNDUR D, HATZINAKOS D. Digital Watermarking Using Multiresolution Wavelet Decomposition［C］. Proc IEEE Int Conference of Acoustics. Speech and Signal Processing, 1998(5):2969-2972.
［4］MICHAEL U. Texture classification and segmentation using wavelet frames［J］. IEEE Transactions on Image Processing, 1995(4):1549-1560.
［5］孙延奎. 小波分析及其应用. 北京：机械工业出版社，2005.
［6］MALLAT S. A theory for multiresolution signal decomposition: the wavelet representation［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1989,11(7):674-693.