Study on the Long-Term Monitoring of Contact Deformation Field of Periodic Rotating Gears Using DIC Combined with SURF Algorithm

Background: Digital image correlation (DIC) is a widely adopted non-contact method for precise motion and deformation measurement, valued for its high accuracy. However, standard 2D DIC struggles to track relative displacements in materials with significant rigid rotations or out-of-plane displacement interference, such as in monitoring rotating gears. Additionally, the extensive data generated during in-situ monitoring makes DIC-based image matching impractical. Objective: This study proposes a method for rapidly identifying identical images from large datasets. The proposed method can also effectively eliminate the impact of rigid rotation and out-of-plane displacement. Methods: This study proposes a dynamic period digital image correlation (DP-DIC) method. The technique utilizes the speeded-up robust features (SURF) algorithm to match and select feature points efficiently, addressing the issue of image decorrelation caused by large-angle rotations. Furthermore, a rigid-body matrix restoration algorithm is incorporated to reduce the effects of rigid rotation and out-of-plane displacement partially. Results: Validation tests for measuring the dynamic deformation field of rotating gears provide essential data. This data supports gear design optimization, performance evaluation, and lifetime prediction. Conclusion: This study proposes a DP-DIC method based on DIC. Validation tests demonstrate that the DP-DIC method is suitable for long-term monitoring of the contact deformation field in periodically rotating gears.