Research on the Innovation and Design Practice of Traditional Batik Based on StyleGAN-Generated Patterns and Eye Movement Signal Optimization

To address the cultural value and design requirements of batik patterns, this article first generates novel patterns incorporating traditional elements such as peonies and phoenixes based on the StyleGAN neural network. Subsequently, eye movement experiments are conducted to screen visually attractive pattern features and evaluate users’ attention distribution toward generated pattern structures. Finally, designers further refine the selected patterns to make them suitable for modern textile design, fabric decoration, and the surface appearance design of functional wearable products. To verify the feasibility of combining the new patterns with contemporary fabrics and engineering-oriented application carriers, design practices are carried out. In particular, the study considers the potential use of patterned textile surfaces in smart fabrics, wearable electromagnetic devices, and antenna-integrated textile products, where visual acceptability and functional integration need to be jointly considered. According to survey results, these designs meet the aesthetic expectations of the audience and show good application potential in textile design, fabric decoration, and the appearance optimization of functional textile products related to wearable electromagnetic applications.