Analysing the Classification of Artistic Styles of Painting in Art Teaching from the Perspective of Emotional Semantics

Abstract

With the advancement of Internet information technology, a great number of art paintings have appeared on key small network platforms. These art paintings include not only a huge quantity of representational information, but also a large amount of semantic information; yet, there is currently a dearth of more systematic research on sentiment semantic analysis in painting art works. To provide basic support for study into the sentiment semantics of art paintings, we present a machine learning-based classification algorithm for painting art in art education from the standpoint of sentiment semantic analysis. To begin, we build machine learning models of different painting styles. Once machine learning is realized, we convert the color space into Lab color space and use the weighting function and the color values of the a and b channels to obtain the image's color entropy; to obtain the chunking entropy, we use the art image's chunking machine learning and the mean of the chunking machine learning; and to obtain the contour Entropy, we use the Contourlet transform to extract the image's contour information. With significant novelty and practical application value, this study presents a new direction for the study and practice of emotional level adaptive interaction in intelligent learning environments.