多介质下三维粒子图像测速算法研究

项目名称： 多介质下三维粒子图像测速算法研究

项目编号： No.61202253

项目类型： 青年科学基金项目

立项/批准年度： 2013

项目学科： 计算机科学学科

项目作者： 杜海

作者单位： 大连理工大学

项目金额： 24万元

中文摘要： 当使用粒子图像测速技术对目标流场进行测量时，在成像系统与示踪粒子间往往存在着两种及两种以上的介质，如空气、玻璃、水等。这些介质不仅改变了粒子成像时的光路而且降低了粒子像的质量，因此测速过程中若不考虑介质的影响便会导致测量结果出现较大的误差。针对这一情况，本项目结合粒子图像测速原理对由介质所引起的折射、散射、吸收等现象进行研究，发现并归纳出介质对粒子图像测速过程的影响规律，并在此基础上提出一种新式的三维粒子图像测速方法，其中包括：粒子退化图像的还原算法、畸变图像的匹配算法以及同名点的对应方法。该套方法的研究综合考虑了粒子图像测速方法自身的影响因素以及系统运行时的外部影响因素，有效地降低或消除了介质的影响，提高了粒子图像测速的准确程度，从而拓展了粒子图像测速技术的应用领域。

中文关键词： 流体测量；粒子图像测速；目标识别；图像匹配；三维重构

英文摘要： When measuring the flow field with three-dimensional particle image velocimetry (3DPIV) technology, there are usually two or more kinds of media between vision system and tracing particles, such as air, glass, water and so on. These media not only change the optical path, but also decrease quality of particle images. The multimedia influence will cause large errors to the measurement results. In order to solve this problem, this research work is carried out: Firstly, the optical phenomena, such as refraction, scattering, absorption caused by the media, will be studied based on the principle of PIV in this project, and the media effect rules will be concluded with PIV experiments. Secondly, a new 3DPIV method will be proposed based on the previous works，which includes restoration algorithm of degenerative particle image, matching algorithm of distortion image and matching algorithm of homonymy points on different images. The method proposed in this project can decrease or eliminate the impact of media and improve the precision of 3DPIV, because the internal factors and external factors of 3DPIV are considered together during investigation on 3DPIV method. So the research of this project is of benefit to the application fields of PIV.

英文关键词： fluid measurement；particle image velocimetry；target identificaiton；image matching；three dimensional reconstruction