Image Expression Classification and Recognition Based on Semantic Knowledge

To reduce the semantic gap between low-level visual features and high-level semantics in images, a method for image expression classification and recognition based on semantic knowledge is proposed. This method utilizes a combination of the Analytic Hierarchy Process (AHP) and the G1-Method to compute high-level semantic knowledge vectors. Further, it incorporates the K-Nearest Neighbor (K-NN) algorithm to integrate low-level visual features and high-level semantic knowledge, thereby narrowing the “semantic gap” between image low-level visual features and high-level semantic knowledge. Experimental results conducted on the BU-3DFE three-dimensional facial expression database demonstrate favorable image expression classification and recognition performance, as well as robustness.