1、摘要 I 摘要 光学测量方法以其非接触、高速度、高精度和全场等特点在现代工业中得到 了广泛的应用。该方法的原理是通过采集和处理光学条纹图像来获取被测信息。 在实际测量中，因为环境的扰动（如振动和温度变化等）和图像设备所固有的局 限性，采集到的条纹图像不可避免地带有噪声，这些噪声也必然会极大地影响测 量精度。 所以在用条纹图像来获取被测信息前， 通常要对条纹图像进行滤波处理。 本论文以计算机图像处理为基础，研究了图像滤波的基本方法，根据条纹图 像的特点，着重研究和实现了两种自适应条纹图像滤波方法：分区域自适应条纹 图滤波方法和基于重心法的方向自适应滤波，并通过 matlab 软件对两种方法进 行
2、了仿真验证。实验结果表明，两种方法能较好地实现条纹图像滤波。最后，论 文还分析实现了基于小波分析的条纹图滤波方法， 并对实验中存在的问题进行分 析，对下一步的研究工作进行展望。 关键词：条纹图像；噪声；滤波；自适应； Abstract II Research and Implementation of the filtering method to fringe image Abstract Optical measurement method for its non-contact, high-speed, high-precision, and audience characterisit
3、ics has been widely used in modern industry. The principle is to obtain the measured information through the collection and processing of optical fringe image. Fringe image inevitable collect the noise in actual measurement because the disturbance of the environment (such as vibration and temperatur
4、e variations, etc.) and the limitation inherent in the image equipment, these noise is bound to greatly affect the measurement accuracy. This thesis is based on computer image processing, studied the basic method of image filtering, focusing on two adaptive fringe pattern filtering method according
5、to the characteristics of the fringe: the subregional adaptive fringe pattern filtering method and the direction adaptive filtering based on the center of gravity. and conducted a simulation of the two methods by matlab software. The experimental results show that both methods can achieve better fri
6、nge image filtering. Finally, the thesis also analyzes the realization of the fringe pattern filtering method based on wavelet analysis, analyze the problem in the experiment, and look out the next research. Key words: Fringe image; Noise; Filtering; Adaptive; 目录 III 目录 摘要 I ABSTRACTABSTRACT . II 第一章 绪论 . 1 1.1 课题研究背景 1 1.2 国内外研究现状 2 1.2.1 国外研究现状 2 1.2.2 国内研究现状 3 1.3 研究内容及论文结构 4 第二章 条纹图像滤波方法概述 5 2.1 条纹图像的特点 5 2.1.1 条纹图像的数学形式与特性 5 2.1.2 实际条纹图的噪声 . 6 2.2 常用条纹图像滤波方法 . 7 2.2.1 线性滤波器 . 8 2.2.2 非
