查看文章

The process to synthesize feature for human facial expression often implies both fuzziness, randomness and their certain relevance in image data. By using the advantage of cloud model, this paper presents a new approaches and applications for comprehensive analysis of human facial expression synthesis using cloud model, in order to realize the rapid and effective facial expression processing in analysis and application. It gives the comprehensive analysis for the fuzziness and randomness of facial expression feature and the relationship between them based on cloud model, including the new method of facial expression synthesis with the uncertainty. It proposes the method of facial expression feature synthesis by cloud model, using the three numerical characteristics (Expectation, Entropy and Hyper Entropy) as the features and concepts of facial expression with its fuzziness, randomness and certain relevance in them. Through such three numerical characteristics, it introduces the framework of facial expression synthesis and the detail procedures based on cloud model. It puts forward the synthesis method of facial expression and gives the concrete realization and the implementation process. The facial expressions after synthesis can express the different expressions for one person, and it can meet a variety of demands for facial expression. The experimental results show that the proposed method is feasible and effective in facial expression synthesis.