Affiliations: Department of Environmental Art, School of Humanities and Arts, Chongqing University of Science & Technology, Chongqing, China
Correspondence:
[*]
Corresponding author. Zhou Yangxiaoxiao, School of Humanities and Arts, Department of Environmental Art, Chongqing
University of Science & Technology, Chongqing, China. E-mail: [email protected].
Abstract: Traditional art design models are inefficient and difficult to control the modal characteristics of images. Based on the actual needs of artistic design, this paper builds an intelligent model of artistic design image modal analysis based on machine learning technology and image recognition technology, and proposes a feature extraction method based on wavelet transform and fuzzy logic and an art image classification method based on rotating quaternion wavelet transform. Moreover, this paper calculates the activation intensity value corresponding to each fuzzy area in the fuzzy feature space and normalizes these activation intensity values to form the artistic image feature vector. In addition, this paper uses rotating quaternary wavelet transform to decompose the art image, and then calculate the energy and standard deviation of each decomposed sub-band coefficient. Finally, this paper uses support vector machines to recognize artistic images. The experimental research shows that the model constructed in this paper has certain effects.
Keywords: Art design, image modality, artificial intelligence, image recognition
