基于压缩感知技术的圆柱度误差全息测量方法研究

The cylindricity is one fundamental geometric tolerance which characterizes the manufacturing performance of workpieces. Among various cylindricity measurement methods, contact measurement method has a shortcoming of low accuracy. Athough non-contact measurement method has an advantage of high accuracy, all point set reflecting the actual geometric tolerances of workpieces can not be sampled by using conventional non-contact methods. Therefore, how to satisfy implicit sampling density and frequency requirements of the GPS sampling standard is needed to be solved, and the above mentioned common and key technology is needed to be paid more attention for noncontact measurement method. The cylindricity error of holographic measurement of cylindricity error based on compressive sensing is proposed in this project. Three dimensional information of the workpieces can be acquired by means of digital holographic microscopy imaging, and data of the holographic image can be compressed and three dimensional images of the workpieces are fast reconstructed by means of compressive sensing. So the holographic image questions of non-uniform speckle noise and huge data are resolved. The principle of holographic measurement of cylindricity error based on compressed sensing can be established. The measurement error can be evaluated according to the GPS specification. Then a novel noncontact and nondestructive cylindricity measurement system with a high accuracy of 0.1µm/mm among 100mm diameter will be developed. The purpose of this project is aimed to set up an accurate cylindricity error measurement and characterization method and provide a precision cylindricity error test and calibration instrument that can be applied in the field of precision machining.

圆柱度是表征轴类零件制造性能的一项基本几何量参数。现有圆柱度误差测量方法中，接触式测量方法测量精度低，非接触式测量方法虽然测量精度高，但是其采样不能获得反映被测零件实际要素的全部点集。因此，如何满足GPS采样规范隐含的采样面密度和采样频率的要求，是圆柱度误差非接触式测量方法需重点解决的关键共性技术。本项目提出基于压缩感知技术的圆柱度误差全息测量方法，以数字全息显微成像获取被测零件圆柱度三维信息，采用压缩感知实现全息图像数据压缩和被测零件的三维快速重建，解决全息再现图像存在非均匀散斑等噪声和记录数据量庞大等问题，建立基于压缩感知的圆柱度误差全息测量原理，依据GPS规范进行测量误差评定，研制一套非接触式圆柱度无损测量系统。拟实现被测轴类零件直径100mm的圆柱度误差测量精度0.1µm/mm。本项目研究旨在建立精确的圆柱度误差测量与表征方法，为精密机械加工提供新型的圆柱度误差测试计量仪器

圆柱度是表征轴类零件制造性能的一项基本几何量参数，圆柱度误差检测技术是支撑高端装备制造技术的一项基础关键技术。近年来，随着高端装备制造技术的高速发展，对轴类零件形状误差的测量精度提出了新的要求。同时，光电检测、机器视觉和信号处理等最新技术，为轴类零件形状误差的高精度测量提供了新的方法和手段。在圆柱度误差测量过程中，满足GPS隐含的采样面密度和采样频率的要求会使得测量数据量激增，难以实现圆柱度误差测量的高效性需求。因此，要求轴类零件的圆柱度误差测量技术及评价方法既要测量精度高又要保证测量的高效性。. 本项目针对激光干涉圆柱面测量技术难点，提出了数字全息和压缩感知技术融合的圆柱度误差测量方法。数字全息记录获取被测圆柱面的三维形貌信息，采用压缩感知压缩提取被测圆柱面点云数据，经数字图像滤波、数字全息再现、坐标变换和圆柱度误差评定过程实现被测圆柱的圆柱度误差测量，实现了亚微米级测量精度，且具有米溯源性。测量系统简单，成本低。在进行圆柱度误差测量过程中，与Rondcom 54dx型圆柱度仪进行相互对比和试验。 . 对平面和凸面圆柱体进行实验测量。实验关键数据如下：. 标准样块：测得粗糙度的测量精度为0.2μm；. 圆柱体：实现被测轴类零件直径30mm和40mm的圆柱度误差测量，圆柱度误差测量精度为0.2μm，仪器示值误差σ=3.0%。实验结果表明本实验装置达到了一级圆柱度仪的技术指标要求。. 科学意义如下：. 1．提出了基于压缩感知技术的大尺寸数字全息图像融合方法，解决了激光光斑尺寸小等因素限制被测物体测量尺寸问题；. 2．鉴于激光干涉在凸面测量过程中一次仅能获取局部测量数据的问题，本文提出了基于数字全息的圆柱度误差测量方法；. 3．提出了基于参照特征的多尺度Harris角点的圆柱面再现像拼接方法，实现各相邻圆柱面全息图的无缝拼接和被测圆柱的三维重建。