基于光照不变特征的数字散斑相关方法

边坡问题是中外工程建设中的主要工程地质问题之一,准确、可靠的边坡变形监测方法对预防滑坡灾害起着至关重要的作用.将数字散斑相关方法应用在边坡变形监测中,可以快速得到准确的边坡位移场信息,具有操作简单、非接触式等优点.但是在野外测量环境下,光照变化对图像的灰度唯一性产生了较大影响,从而导致测量精度的下降.在数字散斑相关方法求解边坡变形场的基础上,使用局部敏感直方图替换原始图像对应位置的灰度值,作为图像的光照不变特征,克服了光照变化对图像灰度信息的影响,并通过数值模拟实验对算法精度进行验证.实验结果表明,该方法大幅提高了边坡变形计算精度,为研究边坡破坏机理和发展过程,提供了一种经济有效的监测方法.

Digital speckle correlation method based on illumination invariant feature

The slope problem is one of the most important engineering geological problems in the engineering construction at home and abroad.The application of digital speckle correlation method in slope deformation monitoring can quickly obtain accurate information of slope displacement field.However, in the field measurement environment, the illumination variation has a great influence on the grayscale uniqueness of the image, which results in the decrease of the measurement accuracy.In this paper, on the basis of solving the slope deformation field with the digital speckle correlation method, local sensitive histogram is used to replace the gray value of the corresponding position of the original image as the illumination invariant feature of the image, overcoming the influence of the illumination change on the image gray information, and verifying the accuracy of the algorithm through numerical simulation experiment.The experimental results show that this method greatly improves the calculation accuracy of slope deformation and provides an economical and effective monitoring method for studying the failure mechanism and development process of slope.