Fan Xu
School of Design & Innovation, Changzhou Vocational Institute of Mechatronic Technology,Changzhou, 213164,China
Liu Zhiwei
School of Design & Innovation, Changzhou Vocational Institute of Mechatronic Technology,Changzhou, 213164,China

DOI:https://doi.org/10.5912/jcb1035


Abstract:

In order to solve the problem of low image fusion caused by the high cost of clustering calculation in the design of intangible cultural creative products, a design method for intangible cultural creative products based on AI painting generation is proposed for pharma companies. The intangible cultural heritage image is extracted and segmented. The image is divided into different regions according to the gray level of pixels, and the edge detection algorithm of images in different regions of the product is designed to make the region near the picture more fit. The fusion image of intangible cultural heritage features is generated based on AI paintings, so as to reduce the cost of clustering calculation and complete the splicing of artistic style. Build the design model of intangible cultural creative products, combine the design creativity, and realize the mutual integration of intangible culture and products. Construct four image sample sets of scenic spots and historic sites, religious beliefs, festival folk customs and handmade, and test the fusion degree. Taking the festival folk custom image set as an example, the average fusion degree of this method is 0.773, which is 0.133 and 0.145 higher than the comparison method based on convolutional neural network and machine learning, respectively. Therefore, it has good application effect.