期刊文献+

基于对比度金字塔图像融合的自发笑脸识别

Spontaneous Smile Recognition Based on CPD Image Fusion
下载PDF
导出
摘要 自发笑脸识别是表情识别研究领域的重要部分,在诸多领域均有着广泛应用。本文提出一种基于对比度金字塔(contrast pyramid decomposition,CPD)图像融合的自发笑脸识别方法。首先将2种光源的图像分别进行4层Gauss塔式分解,建立对应的对比度金字塔,并对每层对应的对比度金字塔采用像素灰度平均法融合规则进行融合,得到融合图像,最后分别提取融合图像的LBP(local binary pattern)特征和LDP(local directional pattern)特征,使用SVM分类器进行分类识别。实验结果表明:使用本文方法融合图像作为识别对象,提取LBP特征和LDP特征的笑脸识别率分别达到96.69%和97.19%,总识别率分别达到96.51%和96.78%,明显优于其他算法,表明本文融合算法提升了自发笑脸识别的识别性能。 Spontaneous smile recognition is an important part in the field of facial expression recognition,which is widely used in many fields.This paper presents a spontaneous smile recognition method by fusing infrared and visible images based on contrast pyramid decomposition(CPD).Infrared and visible images are decomposed with 4layers Gauss Pyramids respectively,and the corresponding contrast pyramids are built.Fusion image which is obtained by combining each layer of infrared image and visible image based on the fusion rule—pixel average.Then LBP features and LDP features are extracted from images,and smile and not smile expressions are classified by SVM classifier.The experimental results show that the smile recognition rates with the fusion images based on LBP features and LDP features can reach 96.69% and 97.19%respectively,and the overall recognition rates with the fusion images based on LBP features and LDP features can reach 96.51% and 96.78% respectively.These results are significantly better than the results with other methods.The results show that the proposed algorithm can improve the recognition performance of the spontaneous smile recognition.
出处 《广西师范大学学报(自然科学版)》 CAS 北大核心 2017年第3期45-52,共8页 Journal of Guangxi Normal University:Natural Science Edition
基金 广西自动检测技术及仪器重点实验室开发基金(YQ1402) 广西多源信息挖掘与安全重点实验室开发基金(MIMS15-05) 广西药用资源化学与药物分子工程重点实验室开发基金(CMEMR2014-B15) 国家星火计划重点项目(2015GA790002)
关键词 图像融合 对比度金字塔分解 笑脸识别 LBP特征 LDP特征 image fusion contrast pyramid decomposition smile recognition LBP feature LDP feature
  • 引文网络
  • 相关文献

参考文献3

二级参考文献80

  • 1王文宁,王汇源.彩色图像中多姿态的人脸快速检测及定位[J].计算机工程,2005,31(17):149-151. 被引量:5
  • 2EKMAN P. Strong evidence for universals in facial expressions: a reply to Russell's mistaken critique[J]. Psychological Bulletin, 1994, 115(2): 268-287.
  • 3MURTHY G R S, JADON R S. Effectiveness of eigenspaces for facial expressions recognition[J]. International Journal of Computer Theory and Engineering, 2009, 1(5): 638-642.
  • 4LI M, YUAN B. A novel statistical linear discriminant analysis for image matrix: two-dimensional Fisherfaces[C]//Proceedings of the 7th International Conference on Signal Processing. Piscataway: IEEE, 2004:1419-1422.
  • 5DEND H-B, JIN L-W, ZHEN L-X, et al. A new facial expression recognition method based on local Gabor filter bank and PCA plus LDA[J]. International Journal of Information Technology, 2005,11(11):86-96.
  • 6SHAN C, GONG S, McOWAN P W. Facial expression recognition based on local binary patterns:a comprehensive study[J]. Image and Vision Computing, 2008,27(6):803-816.
  • 7CAO Y, ZHENG W, ZHAO L, et al. Expression recognition using elastic graph matching[C]//Proceedings of the 1st International Conference on Affective Computing and Intelligent Interaction. Berlin: Springer, 2005:8-15.
  • 8HOMMEL S, HANDMANN U. AAM based continuous facial expression recognition for face image sequences[C]//Proceedings of the 12th International Symposium on Computational Intelligence and Informatics. Piscataway: IEEE, 2011:189-194.
  • 9MISTRY V J, GOYANIM M. A literature survey on facial expression recognition using global features[J]. International Journal of Engineering and Advanced Technology, 2013,2(4):653-657.
  • 10SHARMA A, DUBEY A. Facial expression recognition using virtual neutral image synthesis[EB/OL].[2014-08-07]. http://www.cs.umd.edu/~bhokaal/files/documents/Virtual%20expression.pdf.

共引文献11

;
使用帮助 返回顶部