高应变率变形图像去模糊及数字图像相关测量

Dynamic Deformation Image Deblurring and Image Processing for Digital Imaging Correlation Measurement

针对材料高应变率动态变形中的高速运动模糊图像,提出了一种基于加载分析的点扩散函数去模糊方法,通过对模糊像素进行卷积,有效地去除了试件高速变形过程中所产生的运动模糊。通过霍普金森杆加载实验验证了该方法的有效性,与原始模糊图像的对比表明,该方法可有效地恢复运动模糊图像,从而提高数字图像相关测量的准确性和稳定性。

In this research,we proposed a method for deblurring of images captured in the dynamic deformation of materials. By this method,deblurring was achieved based on a dynamic-based approach, which was used to estimate the point spread function （PSF） during the camera exposure window. The deconvolution process involving the iterative matrix calculations of pixels was then performed on the graphics processing unit （GPU） to reduce the time cost. The proposed method was evaluated using the Hopkinson bar loading system. In comparison with the blurry image, the proposed method successfully restored the image. It is also demonstrated from the image processing applications that the deblurring method can improve the accuracy and the stability of the digital imaging correlation measurement.