Sun Long, Tang Chen, Xu Min, Lei Zhenkun
Appl Opt. 2021 Feb 1;60(4):901-911. doi: 10.1364/AO.403408.
The digital speckle correlation method (DSCM) aims to measure the displacement of the interesting points by matching the subset around the same point between the undeformed image and the deformed image. It is an effective and powerful optical metrology method for deformation measurement. Considering that the gray wolf optimizer (GWO) is one of the most popular metaheuristic algorithms to calculate the unknown search spaces in the field of optical engineering, a sub-pixel displacement measurement technique based on the GWO and gradient algorithm is proposed. First, the zero-mean normalized cross correlation function is applied to analyze the correlation between the reference image and deformed image subsets. Second, by exploiting the global searching ability of the GWO algorithm, the initial integer pixel value is obtained and further viewed as the initialization displacement. Finally, the final sub-pixel displacement is generated by using a Barron gradient algorithm. Compared with the state-of-the-art methods on synthetic speckle images, the proposed method can effectively measure the displacement and deformation of rigid bodies. Furthermore, the experiments on the real images demonstrate the effectiveness of our presented framework.
数字散斑相关方法（DSCM）旨在通过匹配未变形图像和变形图像中同一点周围的子集来测量感兴趣点的位移。它是一种用于变形测量的有效且强大的光学计量方法。考虑到灰狼优化器（GWO）是光学工程领域中计算未知搜索空间最流行的元启发式算法之一，提出了一种基于GWO和梯度算法的亚像素位移测量技术。首先，应用零均值归一化互相关函数来分析参考图像和变形图像子集之间的相关性。其次，利用GWO算法的全局搜索能力，获得初始整数像素值并将其进一步视为初始化位移。最后，使用巴伦梯度算法生成最终的亚像素位移。与合成散斑图像上的现有方法相比，该方法能够有效地测量刚体的位移和变形。此外，在真实图像上的实验证明了我们所提出框架的有效性。
