Research on Wickerwork Patterns Creative Design and Development Based on Style Transfer Technology

Traditional craftsmanship and culture are facing a transformation in modern science technology development, the cultural industry is gradually stepping into digital era, which can realize sustainable development of intangible heritage with help technology. To innovatively generate wickerwork pattern design schemes that meets user’s preferences, this study proposes method patterns based on style migration algorithm. First, an image recognition experiment using residual network (ResNet) convolutional neural applied to Funan establish model. The experimental results illustrate optimal rate 93.37% for entire dataset ResNet50 images, where 89.47%, while traditional 97.14%, 95.95%, personality 90.91%. Second, Cycle-Consistent Adversarial Networks (CycleGAN) build scheme generation models patterns, CycleGAN automatically certain characteristics. Finally, designer uses creative images as inspiration source participates detailed adjustment generated various stylistic features. This proposed could explore application AI providing more comprehensive rich new material creation thus contributing innovation culture. In fact, empower inheritance heritages.