基于KCCA的特征融合方法及人耳人脸多模态识别

鉴于人耳和人脸特殊的生理位置关系，本文从非打扰识别的角度出发，提出一种基于人耳人脸的多模态生物特征识别技术。首先仅采集侧面人脸图像，然后将核方法引入到典型相关分析(CCA)中，提出基于核CCA的特征融合方法，抽取两组特征矢量的非线性典型相关特征构成有效鉴别特征矢量用于识别，并应用其提取人耳人脸的关联特征进行个体的分类识别。实验结果验证了基于KCCA特征融合方法的有效性；此外，与单一的人耳或侧面人脸特征体识别比较，基于人耳和人脸融合的多模态生物特征识别性能得到提高，这为非打扰生物特征识别提供了一条有效的途径。

Feature Fusion Method Based on KCCA and Multimodal Recognition Fusing Ear and Profile Face

A novel multimodal recognition technology based on ear and profile face is proposed. The fusion of ear and face biometrics could fully utilize their connection relationship of physiological location, and possess the advantage of recognizing people without their cooperation. First, only the profile-view face images including ear part were captured for recognition. Then the kernel trick was introduced to canonical correlation analysis (CCA), and the feature fusion method based on Kernel CCA(KCCA) is established. With the method, a kind of nonlinear associated feature was proposed based on ear and profile face for classification and recognition. The results of experiments show that the method is efficient for feature fusion, and ear and profile face based multimodal recognition gets better performance than ear or profile face unimodal biometric recognition. The work provides a new effective approach of non-intrusive biometric recognition.