基于改进Kinect的点分布模型面部表情特征提取  

Feature extraction algorithm for facial expression based on improved Kinect point distribution model

作　　者：曹小峰[1] 程实[1] 陈翔[1] 陈晓红[1] 

机构地区：[1]南通大学计算机科学与技术学院,江苏南通226019

出　　处：《计算机工程与科学》2015年第4期802-805,共4页Computer Engineering & Science

基　　金：国家自然科学基金资助项目(61202006);江苏省自然科学基金资助项目(BK2010280);南通大学自然类基金资助项目(12Z036;11Z071)

摘　　要：在面部表情识别中,差异性会在人体面部充分体现,各个器官的整体分布的复杂性和外界影响等因素,使得人体面部表情特征的定位非常困难。利用Kinect点分布特性,给出了一个面部表情特征提取算法,并进行了相关实验,结果表明,本方法在提取对不相关变形保持不变、噪声不敏感、类别区分度大的特征具有明显优势。In facial expression recognition system, due to the differences of faces,the complexity of organ distribution and the influence of outside factors, facial features location becomes a challenging task. Using Kinect point distribution, we present a feature extraction algorithm for facial expressions, and carry out related experiments. Experimental results show that the proposed method has significant advantages in extracting features that are with unrelated deformation,nonsensitive to noise,or have a high degree of distinction.

关 键 词：KINECT 特征提取 面部表情 情感计算 表情识别 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]