A Face Recognition Algorithm Based on Composite Gradient Vector
-
摘要: 提出了一种新的基于复合梯度向量(Composite gradient vector, CGV)描述的人脸识别算法. 该算法首先在定位后的人脸图像中标定目标区域, 并在目标区域内划分特征子区域, 然后,以特征子区域的边缘奇异点作为向量的起点和终点进行正交采样得到基向量, 将目标区域内所有基向量组建向量簇, 通过对基向量的多维复合得到向量簇内所有极大梯度向量, 最后,以极大梯度向量作为元素组建复合梯度向量并统计复合梯度向量的维度和梯度信息, 将复合梯度向量、复合梯度向量维度和梯度进行人脸库对比, 识别出人脸身份. 该算法抓住了人脸面部特征分散性的特点, 继而对分散性特征采用具有连续性规律约束的复合梯度向量进行描述识别. 实验结果表明, 该算法克服了特征域旋转、光照强度变化及多姿态、多表情对人脸识别的影响, 具有速度快、识别准确、适应性强的特点.Abstract: A novel approach to face recognition based on the composite gradient vector (CGV) is proposed in this paper. Firstly, by detecting the target area information in the located facial image, several image segments are made ready for getting the base vectors with the edges of the sub-area. All the base vectors are made into vector groups. Then the great gradient vector which is extracted from the vector groups constitutes the composite gradient vector. The dimensions and gradient information are calculated for face recognition. Finally, recognition results are obtained by gradient and dimension both of which are extracted from the composite gradient vector. Experiments show that the proposed approach has overcome the feature rotating, varied pose, different facial expressions, and achieves good recognition results.
-
[1] Kirby M, Sirovich L. Application of the Karhunen-Loeve procedure for the characterization of human faces. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1990, 12(1): 103-108[2] Turk M, Pentland A. Eigenfaces for recognition. Journal of Cognitive Neuroscience, 1991, 3(1): 71-86[3] Belhumeur P N, Hespanda J, Kiregeman D. Eigenfaces vs. Fisherfaces: recognition using class specific linear projection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997, 19(7): 711-720[4] Bartlett M S, Movellan J R, Sejnowski T J. Face recognition by independent component analysis. IEEE Transactions on Neural Networks, 2002, 13(6): 1450-1464[5] Moghaddam B. Principal manifolds and probabilistic subspaces for visual recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(6): 780-788[6] Kim K I, Jung K, Kim H J. Face recognition using kernel principal component analysis. IEEE Signal Processing Letters, 2002, 9(2): 40-42[7] Mika S, Ratsch G, Weston J, Scholkopf B, Muller K. Fisher discriminant analysis with kernels. In: Proceedings of the IEEE Signal Processing Workshop. Madison, USA: IEEE, 1999. 41-48[8] Martinez A M. Recognizing imprecisely localized, partially occluded, and expression variant faces from a single sample per class. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(6): 748-763[9] Su Yu, Shan Shi-Guang, Chen Xi-Lin, Gao Wen. Integration of global and local features for face recognition. Journal of Software, 2010, 21(8): 1849-1862(苏煜, 山世光, 陈熙霖, 高文. 基于全局和局部特征集成的人脸识别. 软件学报, 2010, 21(8): 1849-1862)[10] Zhang Xiao-Hua, Shan Shi-Guang, Cao Bo, Gao Wen, Zhou De-Long, Zhao De-Bin. CAS-PEAL: a large-scale Chinese face database and some primary evaluations. Journal of Computer-Aided Design and Computer Graphics, 2005, 17(1): 9-17(张晓华, 山世光, 曹波, 高文, 周德龙, 赵德斌. CAS-PEAL大规模中国人脸图像数据库及其基本评测介绍. 计算机辅助设计与图形学学报, 2005, 17(1): 9-17)[11] Yin Bao-Cai, Sun Yan-Feng, Wang Cheng-Zhang, Ge Yun. BJUT-3D large scale 3D face database and information processing. Journal of Computer Research and Development, 2009, 46(6): 1009-1018(尹宝才, 孙艳丰, 王成章, 盖赟. BJUT-3D三维人脸数据库及其处理技术. 计算机研究与发展, 2009, 46(6): 1009-1018)[12] Tan X Y, Chen S C, Zhou Z H, Zhang F Y. Recognizing partially occluded, expression variant faces from single training image per person with SOM and soft κ -NN ensemble. IEEE Transactions on Neural Networks, 2009, 16(4): 875-886[13] Yang J, Zhang D, Frangi A F, Yang J Y. Two-dimensional PCA: a new approach to appearance based face representation and recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 26(1): 131-137[14] Hu Feng-Song, Zhang Mao-Jun, Zou Bei-Ji, Ma Jun-Rong. Pose and illuminate invariant face recognition algorithm based on HMM with one sample per person. Chinese Journal of Computers, 2009, 32(7): 1424-1433(胡峰松, 张茂军, 邹北骥, 马俊容. 基于HMM的单样本可变光照、姿态人脸识别. 计算机学报, 2009, 32(7): 1424-1433)
点击查看大图
计量
- 文章访问数: 2266
- HTML全文浏览量: 89
- PDF下载量: 978
- 被引次数: 0