Application of Computer Assisted Extraction of Artistic Elements and Illustration Design in Nanyang Han Dynasty Stone Carvings

The goal of this article is to explore the potential of computer-aided technology in the extraction of artistic elements and illustration design of Nanyang Han Dynasty stone reliefs. With the help of constructing and optimizing the convolutional neural network (CNN) model, the rich artistic elements in Nanyang Han stone reliefs can be extracted efficiently and accurately. Comparative experiments were used to evaluate the improvement of model performance under different optimization strategies, especially the influence of weight initialization and over-fitting on model accuracy. The results show that the optimized CNN model performs well in the extraction of artistic elements, improves the accuracy and efficiency of extraction, and retains the cultural connotation and visual characteristics of artistic elements. Furthermore, the extracted artistic elements are applied to illustration design, which shows the possibility and unique charm of combining them with modern design concepts. This application enriches the cultural connotation of illustration works, and also provides a new way for the inheritance and innovation of Han Dynasty stone relief art in Nanyang.