Semantics-Based Art Image Retrieval Using Linguistic Variable

More and more digitized art images are accumulated and expanded in our daily life and techniques need to be established on how to organize and retrieval them. Though content-based image retrieval (CBIR) got great progress, current low-level visual information based retrieval technology does not allow users to retrieval images by high-level semantics. We propose an approach to describe and to extract the fuzzy aesthetic semantics of art images. Accordance with the subjectivity and vagueness of human aesthetic perception, we utilize the linguistic variable to describe the image aesthetic semantics, so it becomes possible to depict the image in linguistic expression such as 'very action'. Furthermore, we apply the feedforward neural network to model the process of human aesthetic perception and to extract the fuzzy semantic feature vector. The retrieval methodology makes users more naturally find desired images by linguistic expression and experimental implementation demonstrates good potential on retrieval art images with a human-accustomed manner.