项目名称: 纳米尺度下的实时三维视觉信息反馈新方法及其关键理论研究
项目编号: No.51205149
项目类型: 青年科学基金项目
立项/批准年度: 2013
项目学科: 机械工程学科
项目作者: 李中伟
作者单位: 华中科技大学
项目金额: 26万元
中文摘要: 纳米尺度下的实时三维视觉信息反馈是实现纳米操作自动化和规模化所需的关键技术。SEM是实现此目标的最佳工具,但现有研究成果只能在特定条件下反馈实时的二维信息或低精度的三维信息,在SEM的成像速度和三维重建算法上还存在诸多基础问题。本项目拟在基于实时视觉跟踪的图像采集策略的基础上,探索性地使用SEM图像序列对操作工具和纳米器件进行实时连续的三维重建,并针对其中涉及的视觉跟踪算法、成像系统模型和参数标定算法等关键问题进行研究。具体包括:提出稳定实时的视觉跟踪方法,并根据其结果制定合理的成像策略,解决成像速度与图像质量之间的矛盾;构建SEM成像系统的通用数学模型和参数标定方法,揭示其真实成像规律。建立参数化的数学模型和标定方法,科学表达其成像规律并简化标定过程。本项目的成功实施将构建一套基于SEM图像序列的纳米尺度下的实时三维视觉信息反馈新方法,为提高纳米操作稳定性和效率提供一种有效的技术手段。
中文关键词: 微纳测量;扫描电子显微镜;通用成像模型;特征检测;三维重建
英文摘要: Real-time 3D visual information feedback at nanoscale is one of key technologies to realize automated nano-handling.SEM is the optimal tool to realize this goal. But real-time 2D information or low precision of 3D information can only be obtained under specific conditions through the existing researches, due to many fundamental problems on the efficiency of imaging and the algorithms of 3D reconstruction. Based on the image acquisition strategy which relies on the real-time visual tracking,this project is planned to reconstruct the operation tools and nanodevice real-time and continuously with exploratory use of SEM image sequences, and to study the key issues on visual tracking algorithm,imaging model,and system calibration method. This project is specified as: to propose a stable and real-time visual tracking method,and then to establish reasonable imaging strategies according to the tracking results to solve the trade-off between imaging velocity and quality; to establish a general imaging model and its corresponding calibration method for SEM helps to reveal the true imaging law of SEM;to establish a parameterized imaging modeling and its corresponding calibration method for SEM in order to scientifically express the imaging rule and simplify the calibration process.The successful implementation of this proj
英文关键词: nano-scale measurement;scanning electron microscope;general imaging model;feature detection;3D reconstruction