Development of Higher-Perthrmance Algorithms for Image Analysis of Time-Resolved PIV

考虑了时间分辨PIV的高阶分析，提出了一种新颖的多帧图像处理算法。速度测量的最终目标是瞬时速度场的识别，而成对图像分析是视频速率测量的最终选择。最新的数字成像设备具有比流量波动更高的采样频率的可能性，它为我们提供了比简单的瞬时速度更多的信息。该算法充分利用了图像相关的全部信息，并给出了速度场的高阶时间导数和空间梯度以及速度本身。为了开发更高性能的层析图像分析算法，本文做了以下研究工作。目前新算法的发展为下一代PIV测量打开了大门，即将推出更高性能的数字图像设备1。回顾和研究了PIV图像分析的数学基础，建立了多帧图像分析的新概念。考虑了基于相关信息的图像分析，提出了高阶分析的几点思路。提出了两种用于多帧PIV图像高阶分析的算法。利用VSJ标准图像进行台架试验，验证了系统的测量性能和精度的提高。

Higher-order analysis of time-resolved PIV was considered, and an innovative new algorithm for multi-frame images has been developed. Final target of the velocity measurement is the identification of instantaneous velocity field, and the pair image analysis was the ultimate choice of the video rate measurement. The latest digital image equipments have the possibility of far higher sampling frequency than flow fluctuation, and it supplies us more information than to obtain simple instantaneous velocity. The algorithm developed in this study enables us to utilize entire information of image correlation, and it gives us the higher-order time-derivatives and spatial-gradients of velocity field together with the velocity itself.In order to develop the higher-performance algorithms for Ply image analysis, the following research works have been done. Present development of new algorithm opens the gate for next generation of PIV measurement with the upcoming higher-performance digital image equipments.1. The mathematical basis of PIV image analysis was reviewed and examined to establish the new concept of multi-frame image analysis. The image analysis based on the correlation information was considered, and several ideas for the higher-order analysis were proposed.2. Two algorithms for the higher-order analysis of multi-frame PIV image were developed. The measurement performances and the improvement of accuracy were verified by means of bench mark test using VSJ standard images.