QQ客服 官方微博 反馈留言

题名	基于SEM的微纳操作系统自动作业技术研究; 基于SEM的微纳操作系统自动作业技术研究
作者	1李勇滔,电工研究所
学位类别	博士
答辩日期	2009-06-02
授予单位	中国科学院电工研究所
导师	1韩立,电工研究所
关键词	微纳操作系统 扫描电子显微镜 自动聚焦 目标识别 三维坐标测定 视觉伺服控制 Micro-Nano Manipulating System Scanning Eletron Microscope(SEM) Auto-focusing Object Recognition 3D Coordination Determination Vision Servoing Control
其他题名	基于SEM的微纳操作系统自动作业技术研究
中文摘要	随着科学技术不断地向微观世界发展，各种微纳操作系统的研究日益活跃。基于SEM的微纳操作系统因其自身的显著优点而逐渐地成为执行微纳操作任务的主要工具。然而，要实现基于SEM的微纳操作系统自动作业，还需要探索和解决的如下关键问题：SEM自动聚焦、基于SEM显微视觉的目标识别、SEM下的三维坐标测定和SEM下操作的控制策略。 这些问题的研究，综合了物理学、数理统计学、图像图形学、模式识别和控制论等多个学科，其中蕴含了大量的的基础研究和应用研究内容。它们的解决，可以大大推进基于SEM的微纳操作系统的自动化和实用化。 首先，本文从SEM显微成像原理和自动聚焦技术的角度，系统地研究了自动聚焦的各种方法。通过比较分析各种SEM显微图像清晰度评价函数和自动聚焦反馈控制算法，提出基于图像分析的SEM自动聚焦方法。该方法的关键在于选择适当的SEM显微图像清晰度评价函数和设计复合式的反馈控制算法。 其次，从图像处理角度研究了SEM显微图像预处理的方法。根据目标识别的基本原理，比较分析了基于模板匹配的识别方法和基于特征匹配的识别方法，并且设计出基于Hu不变矩的SEM下目标识别算法；考虑到SEM的自身环境，初步建立了在SEM下静止目标和运动目标的三维坐标测定方法，设计出基于规范化谱主元分析的深度信息提取方法来获取SEM下运动目标的深度信息。 再次，本文介绍了视觉伺服控制的基本原理和分类，比较分析基于位置的视觉伺服控制、基于图像的视觉伺服控制和学习控制等三种控制策略。结合SEM显微视觉的特点，初步提出基于CMAC学习控制的SEM显微视觉伺服控制策略。 最后，本文采用环境扫描电子显微镜作为微纳操作系统的观测器，利用高精度的纳米操作臂作为微纳操作器，构建了基于ESEM的微纳操作系统。总结了课题研究的主要工作内容，并对该技术领域的有关问题进行了分析和展望。 With the continuous development of science and technology in a microscopic world, various kinds of micro-nano manipulating system have been researched and designed in these years. Because of its remarkable advantages, SEM based micro-nano manipulating system has been the main tool implementing micro-nano scale manipulation tasks. However, in order to accomplish automatic working using SEM based micro-nano manipulating system, some critical issues need to analyse and solve, such as SEM auto-focusing, SEM micro-vision based object recognition, 3D coordination determination on SEM, and control strategy in manipulating on SEM. The research on these problems covers many subjects, for example physics, mathematical statistics, image and graphics, pattern recognition, and cybernetics, involving a great deal of content about fundamental research and applied research. These problems-solving can promote the automation and practical application greatly. First, in view of the SEM micro-imaging principle and auto-focusing technology, different kinds of auto-focusing methods have been studied systematically. By analysing comparatively variety of SEM micro-image sharpness evaluation functions and auto-focusing feedback control algorithm, an automatic focusing technology of SEM image-vision based on image analysis is presented. The key point of this technology lies in the selection of suitable SEM micro-image sharpness evaluation functions and the design of compound feedback control algorithm. Second, from the perspectives of image processing, SEM micro-image preprocessing technology has been studied. According to object recognition basic principle, template-matching based recognition method and feature-matching based recognition method are analysed comparatively, and object on SEM recognition algorithm using Hu moment invariants is designed. Considering the SEM environment, 3D coordination determination technology of stationary target and motion target on SEM is initially established. The depth information extraction using normalized spectral principal component analysis is presented to acquire the depth information of motion target on SEM. Third, the main principle and classification of vision servoing control are introduced in detail. By means of the comparative analysis of location-based vision servoing control, image-based vision servoing control and learning control based vision servoing control, in combination of SEM micro-vision characteristic, CMAC（Cerebellar Model Articulation Controller）learning control based SEM micro-vision servoing control method is proposed tentatively. Finally, ESEM based Micro-Nano manipulating system has been constructed, in which ESEM（Environmental Scanning Electron Microscope）is used as the observer of micro-nano manipulating system, and high-precision nanobar is used as micro-nano manipulator. The main work of the present research is concluded and the prospect of this technology is introduced.
语种	中文
公开日期	2010-10-18
页码	129
分类号	TM1;TM921
内容类型	学位论文
源URL	[http://ir.iee.ac.cn/handle/311042/6778]
专题	电工研究所_其他部门_其他部门_博士学位论文
推荐引用方式 GB/T 7714	1李勇滔,电工研究所. 基于SEM的微纳操作系统自动作业技术研究, 基于SEM的微纳操作系统自动作业技术研究[D]. 中国科学院电工研究所. 2009.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。