Design and Application of Artificial Neural Network Algorithm for Digital Speckle Correlation Method

This study explores the design and application of an Artificial Neural Network (ANN) algorithm for Digital Speckle Correlation Method (DSCM). DSCM, as a non-contact full-field deformation measurement technique, has significant applications in materials science and structural engineering. However, traditional DSCM has limitations in handling complex deformations and noise interference. To overcome these challenges, we propose an ANN-DSCM algorithm based on Convolutional Neural Networks (CNN). The algorithm comprises four main modules: feature extraction, correlation calculation, displacement estimation, and refinement, capable of learning and predicting deformation fields directly from speckle image pairs.We constructed comprehensive training and testing datasets using synthetic and experimental data, covering various deformation modes and image quality conditions. The network was trained using supervised learning methods, and its performance was validated using multiple evaluation metrics. Results show that ANN-DSCM demonstrates higher accuracy and robustness compared to traditional DSCM methods in handling large deformations, discontinuous deformations, and noise interference. Moreover, ANN-DSCM exhibits advantages in computational efficiency due to its parallel computing capabilities.This research not only advances DSCM technology but also provides new insights into applying deep learning in experimental mechanics. Future work will focus on further optimizing network structures, expanding application ranges, and exploring the potential applications of ANN-DSCM in material characterization and structural health monitoring.