English

应用科学学报 ›› 2009, Vol. 27 ›› Issue (6): 574-579.

• 信号与信息处理 • 上一篇    下一篇

利用数字图像相关性实现岩土自动跟踪

王伟1;2， 何小元2， 孟云梅1   

  1.  1．南京工程学院建筑工程学院，南京211167
    2．东南大学工程力学系，南京210096
  • 收稿日期:2009-07-24 修回日期:2009-10-21 出版日期:2009-11-25 发布日期:2009-11-30
  • 作者简介:王伟，博士，讲师，研究方向：实验力学的测试和分析，E-mail: ww1177114@163.com；何小元，教授，博导，研究方向：光测力 学，E-mail: mmhxy@seu.edu.cn
  • 基金资助:

    南京工程学院科研基金(No.KXJ08016)资助项目

Automatic Tracking of Soil Using Digital Image Correlation

WANG Wei1;2, HE Xiao-yuan2, MENG Yun-mei1   

  1. 1．College of Architecture Engineering, Nanjing Institute of Technology, Nanjing 211167, China
    2．Department of Engineering Mechanics, Southeast University, Nanjing 210096, China
  • Received:2009-07-24 Revised:2009-10-21 Online:2009-11-25 Published:2009-11-30

PDF

1439

被引次数

1

摘要：

为了实时了解岩土微细结构动态演变规律，针对土颗粒在载荷下的移动特点，研究了适于岩土自动跟踪系统
的图像相关性计算公式、配准模板、搜索区域. 通过比较4 种相关系数的计算耗时、峰度、统计分布，提出在相关性计算
公式中用平方运算代替一般的开方运算，并将正方形模板和正方形搜索区域改为矩形模板和矩形搜索区域，充分利用模
板的灰度信息，满足了岩土跟踪的实时性要求. 实验表明该方法可靠，可为岩土微细结构连续变化特性研究提供实验
手段.

关键词: 数字图像相关, 自动跟踪, 岩土, 连续变化, 相关函数

Abstract:

In order to explore the dynamic evolution law of soil microstructures in real time, image correlation,
matching template and search region adapted to the automatic tracking system are studied towards moving features
of soil particles under loadings. It is vital to satisfy the real-time requirement and accuracy performance of the
tracking system. By analyzing calculation time, kurtosis and statistical distribution of four types of correlation
coefficients, we propose to use the square relation instead of the traditional cubic formula. The rectangular template
and search region are used to replace the square template and search region to take advantage of the template’s gray
information and meet the real-time requirement. Experimental results show that the automatic tracking system is
reliable, which provides an experimental means for the research of continuous deformation in the soil microstructure.

Key words: digital image correlation, automatic tracking, soil, continuous deformation, correlation function

中图分类号: