应用于微靶装配的显微视觉检测技术  

Micro-vision measuring technology in microtarget assembling

作　　者：吴文荣[1] 吴倩[1] 黄燕华[1] 

机构地区：[1]中国工程物理研究院激光聚变研究中心,四川绵阳621900

出　　处：《中国光学与应用光学》2008年第1期75-79,共5页Chinese Optics and Applied Optics Abstracts

基　　金：国家863高技术研究发展计划探索基金资助项目(No.2007AA804401)

摘　　要：针对如何实现微小尺度(μm^mm)靶零件装配过程中的在线检测,研制了应用于微靶装配的显微视觉检测系统,并对系统的构成与显微成像、图像预处理、特征检测等关键图像处理技术进行了详细论述。提出了针对微靶零件的新型多重滤波除噪算法和几何特征检测算法,并以铝台阶、柱腔等微靶零件为应用实例开展了实验研究。结果表明,该系统采用的算法能有效、快速、准确地对靶零件进行几何特征和实时位置检测,在检测视场为3 mm时,其检测精度≤3μm,角度检测精度≤0.1°,适用于ICF微靶等微型器件的微装配。To solve the problem of on-line measuring micro-target parts in assembling process, a micro-vision measuring system was designed and developed in this paper. The composition of miro-vision system, implementation of micro image capturing, image preprocessing, and character measurement were described. In the pre-processing, a new refihering and denoising algorithm and a character measuring algorithm were presented for micro-target parts. Then, the system was used to measure the microtarget parts such as A1 sidestep sample, and hohlraums. Experimental results indicate that proposed system can measure the geometric characters and sample positions in real time, high speed and high precision for target parts. When measuring field of view is 3 mm, the geometric measuring precision and angle precision are less than 3 μm and 0. 1 °, respectively, which shows proposed system is suitable for the assembling process of micro-target part.

关 键 词：微装配 显微视觉 图像预处理 特征检测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]