Optimization of SVM parameters in gait recognition using improved grey wolf algorithm

The recognition accuracy of gait using Support Vector Machines (SVM) is significantly influenced by hyperparameters, while traditional optimization methods are prone to local optima. This study proposes an Improved Grey Wolf Optimization algorithm (IGWO) integrated with a Dynamic Learning-based Hunting (DLH) strategy to optimize SVM parameters. The experiment shows that the proposed framework provides an efficient solution for real-time gait recognition in wearable systems, establishing an advanced optimization paradigm for biomechanical signal analysis.