摘要
在数字图像相关法应用中,与传统散斑制作方法相比,参数化制作方法可以控制散斑的特征参数,确保散斑质量,避免对操作者经验的过度依赖。为了确定散斑特征参数的最佳取值范围,首先生成一系列具有不同特征参数的仿真散斑图,然后对其施加亚像素位移,以位移计算误差作为评价标准,来比较散斑特征参数对图像匹配精度的影响。为了使仿真散斑图接近真实成像,利用高斯滤波对二值仿真散斑图进行预处理,用于模拟实际散斑成像中斑点边缘灰度过渡的效果,以拟合误差与斑点中心灰度综合评价高斯滤波参数对仿真效果的影响,具体分析了高斯滤波参数选择对仿真效果的影响。
In the application of Digital Image Correlation(DIC),compared to traditional speckle fabrication methods,parametric fabrication methods allow for the control of speckle characteristic parameters,ensuring speckle quality and reducing excessive reliance on the operator's experience.To determine the optimal range of speckle characteristic parameters,a series of simulated speckle patterns with different characteristic parameters are first generated.Subpixel displacements are then applied to these patterns,and the displacement calculation error is used as an evaluation criterion to compare the impact of speckle characteristic parameters on image matching accuracy.In order to make the simulated speckle pattern closer to real imaging,Gaussian filter is used to preprocess the binary simulated speckle pattern,which is used to simulate the gray transition effect of spot edges in actual speckle imaging.The influence of Gaussian filter parameters on the simulation effect is comprehensively evaluated by fitting error and spot center gray level.The specific impact of Gaussian filterg parameter selection on the simulation effect is analyzed.
作者
梁皓
付建
于海洋
王文杰
Liang Hao;Fu Jian;Yu Haiyang;Wang Wenjie(Naval Submarine Academy,Qingdao,China)
出处
《科学技术创新》
2025年第17期33-37,共5页
Scientific and Technological Innovation
关键词
数字图像相关法
参数化散斑
参数优化
高斯滤波
digital image correlation
parametric speckle
parameter optimization
Gaussian filter
