Autonomous Classification System for Digital Painting Images Based on Artificial Intelligence

With the innovation of computer intelligent vision technology, the target requirements for image classification are constantly improving, and the design and research of digital painting image (DPI) autonomous classification optimization system based on CNN (Convolutional Neural Networks) are becoming increasingly important. In the construction of the entire DPI autonomous classification system, how to improve the classification accuracy and feature extraction efficiency of the improved system is currently a key issue that needs to be urgently solved. In this paper, through the relevant research on the traditional autonomous classification methods of DPI, with the help of CNN in the application of digital painting image classification and algorithm steps, combined with the digital image normalization and equalization processing calculation, the following conclusions were drawn based on the data results. Six digital painting image samples were selected through simulation experiments, and the application of CNN based DPI autonomous classification system improved the classification accuracy compared to traditional schemes, with a comprehensive average improvement of 5.3%. At the same time, there were also improvements in feature extraction, with a comprehensive average improvement of about 10%. This study provides an innovative method that enables computers to independently understand and classify various digital painting works through deep learning and image recognition technology, providing effective tools for art management, collection, and display, and expanding the application prospects of machine vision in the field of culture and art.