基于小波多尺度分析的PIV互相关算法研究  

Study on algorithms of PIV with wavelet multi- resolution analysis

作　　者：金上海[1,2] 陈刚[2] 许联锋[2] 邵建斌[2] 李志高[2] 

机构地区：[1]西安理工大学应用数学系,陕西西安710048 [2]西安理工大学水力学研究所

出　　处：《水动力学研究与进展（A辑）》2005年第3期373-380,共8页Chinese Journal of Hydrodynamics

摘　　要：随着计算机和图像处理方法的迅猛发展,粒子图像测速仪(ParticleImageVelocimetry简称PIV)和图像测量技术得到了迅速的发展。本文研究了基本的互相关算法和基于快速傅立叶变换的互相关算法,以图像互相关理论为基础,借鉴小波变换和多尺度分析的思想,设计了一个分层匹配算法,克服了傅立叶变换不易提供信号完全信息的缺陷。最后,在VIS—UALFORTRAN环境下实现了算法,处理模拟粒子图像的结果与模拟数据比较吻合。将该算法应用于实际拍摄的水垫塘中固液两相流图像,也收到了良好效果。Particle Image Velocimetry (PIV) and its image measurement techniques have heen rapidly developed with the drastic development of image processing method and computers. The conventional cross-correlation algorithm and the cross-correlation algorithm based on Fast Fourier Transform are discussed. To conquer the limitation of Fourier Transform, a delam-inated processing method based on the theory of cross-correlation of image, used for reference of wavelet transform and multi-scale analysis, is presented in this paper. The realization procedures of the method via VISUAL FORTRAN are provided. The computational results of the simulated particle images are in reasonable agreement with the simulated data. It is applied to measurements of actual flows, and the effect is better.

关 键 词：图像测量 粒子图像测速仪 多尺度分析 小波变换 

分 类 号：TV135[水利工程—水力学及河流动力学]