Investigation of the Relationship Between the Velocity of Magnetic Domain Wall and Magnetic Barkhausen Noise Using Classic + NL Algorithm

Domain wall (DW) velocity is a key parameter for understanding magnetic domain dynamics in ferromagnetic materials. However, traditional algorithms for DW velocity extraction often suffer from inaccuracies, limiting the precise correlation between microscopic domain dynamics and macroscopic magnetic Barkhausen noise (MBN). In this study, we quantitatively compare the accuracy of multiple optical flow algorithms in DW velocity extraction. First, we establish rigorous error metrics to evaluate optical flow performance. We then optimize the parameters of three optical flow algorithms—Horn–Schunck (H–S), Brox, and Classic + NL—and demonstrate that Classic + NL outperforms the others in both standard test images and experimental domain motion images. Furthermore, we validate the Classic + NL algorithm’s reliability through correlation analysis between the DW velocity and MBN envelope. This framework provides a robust foundation for linking microscopic DW motion with macroscopic MBN responses, advancing the precision of domain dynamics characterization.