Yangzhou lacquer art, as a precious historical and cultural heritage in China, has a high reference value for modern tea set design. In this paper, firstly, the development and design method of the tea set based on Yangzhou Lacquer Art is combined with SIFT and BRISK feature extraction algorithms to screen out the stable image feature points and feature extract the traditional decorative patterns of the tea set, and then the neighborhood pixel gradient distribution of the feature points is counted through the histogram to determine the maximum peak of the neighborhood gradient. Subsequently, the RANSAC algorithm is applied to calculate the coordinate transformation relationship between the feature points of the template image and the corresponding feature points of the image to be matched, and a portion of the matching points with large errors are removed to obtain the correct matching points. Using a Generative Adversarial Network, the efficient hyper-segmented pattern tattoos are reconstructed, and the RMGAN loss function is designed to satisfy the problem of stochastic gradient descent of the image, and the generated image is optimized step by step. Through surface modeling to assist Yangzhou lacquer art tea set shape design modeling and the local shape of the tea set, to realize the physical design of the tea set of Yangzhou lacquer art. Empirical investigation on the market feedback after the design of the tea set, from the results of the survey: different groups of people on the use of the process of satisfying behavior, sensory experience demand is high, especially the youth group, the degree of demand is 0.7456, 0.7655. Users want to use classical colors and traditional painting elements in the design of the tea set, with a response rate of 23.91% and 30.43%, respectively. Combining the technical method developed in this paper with the results of demand analysis, it provides a practical method for applying Yangzhou lacquer art to tea set design.