数字图像相关中基于非线性灰度改变模型的灰度梯度迭代算法  

Gray-gradient Iterative Algorithm Based on Nonlinear Intensity Change Model in Digital Image Correlation

作　　者：熊磊[1] 刘国栋[1] 刘小勇[1,2] 杨鑫[1] 刘健英[1] 谭庆昌[1] 

机构地区：[1]吉林大学机械科学与工程学院,吉林长春130022 [2]北华大学机械工程学院,吉林吉林132021

出　　处：《机床与液压》2012年第13期70-72,共3页Machine Tool & Hydraulics

摘　　要：传统的数字图像相关方法用于位移及应变场的测量都是基于物体表面同一点在变形前后灰度不变的基本假设。但是在实际应用中,灰度会受到光照及环境等因素的影响而随着时间发生变化。针对这一问题,将比较符合实际情况的非线性灰度改变模型引入到灰度梯度迭代算法中,并通过计算机模拟实验验证算法的精度及稳定性。结果表明,改进后的算法能够降低由于灰度改变所引起的测量误差。In traditional digital image correlation method for displacement and strain measurements, the fundamental assumption of the method is the gray level intensity of a physical point on object surface does not change before and after deformation. However, the intensity in practical applications may change at various time due to the illumination and environment condition. A more practical nonlinear model for intensity change was used in gray-gradient iterative algorithm. The accuracy and stability of the algorithm were ver- ified through computer simulated experiments. The results show that the improved algorithm ean reduce measure error caused by inten- sity variation.

关 键 词：数字图像相关 非线性灰度改变模型 灰度梯度迭代算法 

分 类 号：O348[理学—固体力学]