Microscopic DIC measurement using an improved global registration image-stitching algorithm

In this study, an improved microscopic digital image correlation (DIC) measurement method using a global registration image-stitching algorithm is proposed to address the challenge of balancing strain measurement accuracy with computational efficiency in large-field microscopic measurements. First, multiple adjacent images are captured at different locations of microscopic structures. Subsequently, the captured images are stitched together using an improved global registration optimization algorithm to obtain the pre- and post-deformation stitched images. Finally, the DIC method is applied to compute the deformation field. Furthermore, the feasibility of the proposed method is first verified through image-stitching experiments and rigid-body translation simulations. Subsequently, the method's effectiveness in real-world deformation measurements is further validated through rigid-body translation and uniaxial compression experiments. Experimental results show that the proposed method improves the average structural similarity index by 27% and reduces stitching time by 16% compared with traditional methods. For displacements of as small as 0.1 pixel, the average error and standard deviation of the stitched images are controlled within 0.02 pixels. In practical measurements, the micrometer-level strain error is maintained within 20 µε. These findings confirm that the proposed method considerably extends the DIC measurement range while ensuring accuracy and efficiency.