数字图像相关方法中基于熵的散斑子图质量评价研究

在数字图像相关测量计算过程中,直接参与互相关计算的是散斑子图,而散斑子图质量对测量精度具有重要影响。因此,如何有效评价散斑子图质量是数字图像相关方法研究的热点问题。将信息论中用于衡量信息量的“熵”概念引入散斑子图评价,提出以图像熵为评价参数的散斑子图质量评价方法。为验证该评价参数的有效性,对采用不同方法采集的3幅不同类型散斑图进行位移测量实验。实验结果表明:位移测量精度与散斑子图熵有紧密的关系,散斑子图熵越大,位移测量精度越高。实验结果验证了图像熵可有效用于散斑子图的质量评价。

Quality Assessment of Speckle Subimage Based on Entropy in Digital Image Correlation Method

In the process of digital image correlation measurement calculation,the speckle subimages are directly used for cross-correlation calculation,and the quality of speckle subimages has important effect on the measurement accuracy.Therefore,how to effectively assess the quality of speckle subimage is a point issue in the study of digital image correlation method.The concept of entropy,which was used to measure the amount of information in information theory,was introduced into the assessment of speckle subimage,and then a new method for quality assessment of speckle subimage based on image entropy was proposed.In order to verify the validity of the assessment parameter,the displacement measurement experiments were carried out on three different types of speckle subimage obtained by different methods.The experimental results show that the accuracy of displacement measurement is closely related to the speckle subimage entropy,the bigger the entropy of speckle subimage is,the higher the accuracy of displacement measurement is.The experimental results verify that the image entropy can be used to assess the quality of speckle subimage effectively.