Research On Emotional Classification Method Of Chinese Paintings

Emotional is a kind of ability that is peculiar to people or some advanced intelligent creatures.Emotional analysis can be regarded as an important symbol of artificial intelligence.Vision is an important way for people to perceive the world and to obtain information.And human vision includes the images and the scenes.Therefore,the image is an important medium to transmit information.The image itself contains a huge amount of information,including multi-level semantic information.We generally think that the image has three semantic levels: feature semantics,content semantics and emotional semantics.Semantic gap exists between the levels,and it is difficult to map from the bottom to the top.Chinese paintings are the art treasures of our country.Traditional Chinese paintings belong to an emotion-centered art form.Because of importance of emotion in traditional Chinese painting,we must inevitably consider the emotional factors in the appreciation,research and protection of Chinese painting.This paper focuses on image feature extraction,semantic extraction,classification methods and so on in image emotion semantic analysis.We extract colors,textures,regions of interest and spatial features and use PCA to reduce the feature dimension,fuse the features and normalize,and finally classify.We not only concerned with the traditional feature extraction and mapping methods,but also try to use the convolution neural network to make it possible to effectively bridge the semantic gap of traditional Chinese painting,and we try to explore the connection between the underlying features and sentiment,and establish the accurate mapping relationship between the emotional features and the image features to improve the accuracy of emotional recognition classification of Chinese painting images.We can also enrich the organization,management and retrieval methods of traditional Chinese painting and popularize the spread of traditional Chinese painting while appreciating and studying traditional Chinese paintings.