Abstract:

In order to improve the accuracy of facial fine feature recognition， a facial fine feature recognition algorithm considering the correlation between learning interests and micro expressions is proposed. Selecting image entropy as the objective function for facial image segmentation， using particle swarm optimization （PSO） algorithm to optimize the parameters of pulse coupled neural network （PCNN）， determining the optimal values of key parameters， achieving facial image segmentation， and identifying key areas such as eyes and mouth. On the basis of analyzing the correlation between learning interests and micro expressions， the Harris algorithm is used to filter the scale invariant feature transform （SIFT） feature points， accurately locking the key interest points in facial expression images. A strategy based on the maximum coverage area and its adjacent range of feature points to capture the features of each region. The filtered area is used as input for local binary patterns （LBP） feature extraction， and facial fine feature recognition is achieved through support vector machine （SVM） multi classification technology. The experimental results show that the proposed algorithm performs well in facial image segmentation and has high accuracy in recognizing subtle facial features.

Key words: learning interests, microexpressions, relevance, fine facial features, recognition