基于PSO与ICA的表情特征提取

提出基于粒子群优化（PSO）与独立分量分析（ICA）的表情特征提取方法。首先利用ICA算法对表情图像数据建立基本的独立基向量求解框架；为了减少计算复杂度，然后利用PSO算法对处理后的表情图像数据搜索最优的解集合；最后利用支持向量机（SVM）作为算法验证的分类器。实验结果表明该算法在保证较高表情识别率的基础上加快了表情图像特征提取的速度。

Expression feature extraction based on PSO and ICA

A combined method of expression feature extraction was proposed based on particle swarm optimization (PSO) and independent component analysis (ICA). Firstly, the basic ICA algorithm was used to build a solving frame of the independent base vector derived from expression images. Secondly, the PSO algorithm was applied to process expression data to get the optimal solution in order to decrease the computing complexity. Finally, Support Vector Machine (SVM) was used to validate the correctness of algorithm. The experiments in the expression database show that it is a faster way of extracting expression features without sacrificing the expression recognition precision.