| DIGITAL SPECKLE CORRELATION METHOD IMPROVED BY GENETIC ALGORITHM |
| |
| 作者姓名： | MaShaopeng JillGuanchang |
| |
| 作者单位： | [1]DepartmentofEngineeringMechanics,TsinghuaUniversity,Beijing100084,China [2]DepartmentofGeophysics,PekingUniversity,Beijing100871,China |
| |
| 基金项目： | Project supported by the National Natural Science Foundation of China(No.19772033)，the Research Innovation Fund of Tsinghua University for Ph.D.Candidates(No.092410048). |
| |
| 摘 要： | The digital speckle correlation method is an important optical metrology for surface displacement and strain measurement. With this technique, the whole field deformation information can be obtained by tracking the geometric points on the speckle images based on a correlation-matching search technique. However, general search techniques suffer from great computational complexity in the processing of speckle images with large deformation and the large random errors in the processing of images of bad quality. In this paper, an advanced approach based on genetic algorithms (GA) for correlation-matching search is developed. Benefiting from the abilities of global optimum and parallelism searching of GA, this new approach can complete the correlation-matching search with less computational consumption and at high accuracy. Two experimental results from the simulated speckle images have proved the efficiency of the new approach.
|
| 关 键 词： | 数字点相关理论 遗传算法 变形测量 应变测量 固体机械 断裂理论 固体力学 |
| 收稿时间： | 8 July 2002 |
DIGITAL SPECKLE CORRELATION METHOD IMPROVED BY GENETIC ALGORITHM |
| |
| MaShaopeng JillGuanchang.DIGITAL SPECKLE CORRELATION METHOD IMPROVED BY GENETIC ALGORITHM[J].Acta Mechanica Solida Sinica,2003,16(4):366-373. |
| |
| Authors: | Ma Shaopeng |
| |
| Institution: | (1) Department of Engineering Mechanics, Tsinghua University, 100084 Beijing, China;(2) Department of Geophysics, Peking University, 100871 Beijing, China |
| |
| Abstract: | The digital speckle correlation method is an important optical metrology for sur- face displacement and strain measurement.With this technique,the whole field deformation in- formation can be obtained by tracking the geometric points on the speckle images based on a correlation-matching search technique.However,general search techniques suffer from great com- putational complexity in the processing of speckle images with large deformation and the large random errors in the processing of images of bad quality.In this paper,an advanced approach based on genetic algorithms (GA) for correlation-matching search is developed.Benefiting from the abilities of global optimum and parallelism searching of GA,this new approach can complete the correlation-matching search with less computational consumption and at high accuracy.Two experimental results from the simulated speckle images have proved the efficiency of the new approach. |
| |
| Keywords: | digital speckle correlation method genetic algorithm deformation measurement |
| 本文献已被 CNKI 维普 SpringerLink 等数据库收录！ |
