Incremental digital volume correlation method with nearest subvolume offset: An accurate and simple approach for large deformation measurement. (February 2018)

• Record Type: Journal Article
• Authors: Wang, Bo; Pan, Bing
• Is Part Of: Advances in engineering software. Volume 116(2018)
• Page Start: 80
• Page End: 88
• Publication Date: 2018-02
• Subjects: Digital volume correlation -- Large deformation -- Nearest subvolume offset; Computer-aided engineering -- Periodicals; Engineering -- Computer programs -- Periodicals; Engineering -- Software -- Periodicals; Periodicals; 620.0028553
• DOI: 10.1016/j.advengsoft.2017.12.004
• Languages: English
• ISSNs: 0965-9978

Abstract

Highlights: An accurate and simple incremental DVC is proposed for large deformation measurement. The reference subvolumes in the updated reference volume images are translated to nearest integer-voxel positions. Subvoxel intensity interpolation for the updated reference subvolumes can be entirely avoided. The proposed incremental DVC outperforms existing ones in terms of accuracy and efficiency. Abstract: Digital volume correlation (DVC) has been widely accepted as an effective experimental technique for quantifying full-field internal 3D deformation of solid materials and structures under external loading. However, conventional DVC using a fixed reference volume image generally fails when serious decorrelation occurs in deformed volume images due to large deformation or other reasons. In this work, an accurate and simple incremental DVC method with nearest subvolume offset is proposed for large deformation measurement. Specifically, the reference subvolumes in the updated reference volume images are translated to nearest integer-voxel positions, rather than being interpolated at subvoxel locations. The translated reference subvolumes, within which all correlation points locate at integer-voxel positions, are then tracked in the rest deformed volume images to retrieve incremental displacement fields. The obtained incremental displacement fields are then accumulated to previously obtained displacement fields to determine the overall displacements. By using the simple nearest subvolume offset approach, subvoxel intensity interpolation for the updated reference subvolumes is entirely avoided, thus not only eliminating the bias error associated with imperfect subvoxel intensity interpolation, but also increasing the computational efficiency of incremental DVC calculation by approximately 2.5 times. The accuracy, efficiency and practicality of the presented incremental DVC are demonstrated by analyzing two sets of volume images with large deformation generated in numerically simulated and real-world experiments.