自然场景下人脸表情数据集的构建
|
摘要
目前人脸识别研究中表情数据集图像数量较少、表情信息单一,不利于人脸表情识别的研究。本文创建了自然场景下带标签的人脸表情数据集(Facial expression dataset in the wild,FELW),并对其进行测试。FELW表情数据集包含了多张从互联网上收集的不同的年龄、种族、性别的人脸表情图像,采用适合的方法标注每张图像带有人脸部件的状态标签和表情标签,并引入Kappa一致性检验,提高人脸表情识别率。使用传统方法和深度学习的表情识别方法对数据集进行实验分析,与其他公开的人脸表情数据集相比,FELW数据集具有更多图像和更丰富的表情类别,并包含了两种图像标签有利于表情识别的研究。
Nowadays,there are many facial expression datasets for expression research. But,these datasets are small amounts of images and little expression information,which limits the expression research. This paper introduces the construction of the facial expression dataset in the wild(FELW)and the test case. The FELW dataset includes many facial expression images,and the images of different age,race and gender are collected from the Internet. Each image includes two labels—the state of facial part label and the expression label,that labeled by the appropriate method. And the author imported Kappa consistency check to improve the recognition rate of the FELW dataset. The author used traditional method and deep learning method to experiment and analysis on this dataset. Compared with other public facial expression datasets, the FELW dataset has much more amounts of images and more varieties of expression,and contains two labels to help to the expression research.
Nowadays,there are many facial expression datasets for expression research. But,these datasets are small amounts of images and little expression information,which limits the expression research. This paper introduces the construction of the facial expression dataset in the wild(FELW)and the test case. The FELW dataset includes many facial expression images,and the images of different age,race and gender are collected from the Internet. Each image includes two labels—the state of facial part label and the expression label,that labeled by the appropriate method. And the author imported Kappa consistency check to improve the recognition rate of the FELW dataset. The author used traditional method and deep learning method to experiment and analysis on this dataset. Compared with other public facial expression datasets, the FELW dataset has much more amounts of images and more varieties of expression,and contains two labels to help to the expression research.
引文
[1]周飞燕,金林鹏,董军.卷积神经网络研究综述[J].计算机学报,2017,40(6):1229-1251.Zhou Feiyan,Jin Linpeng,Dong Jun.Review of convolutional neural network[J].Chinese Journal of Computers,2017,40(6):1229-1251.
[2]甘荧.自然场景下人脸表情数据集的构建与应用研究[D].南昌:江西师范大学,2016.Gan Ying.Construction and application research of facial expression datasets in wild[D].Nanchang:Jiangxi Normal University,2016.
[3]华琳,阎岩,张建.关于对诊断一致性Kappa系统的探讨[J].数理医药学杂志,2006(5):518-520.Hua Lin,Yan Yan,Zhang Jian.Discussion on diagnostic consistency Kappa system[J].Journal of Mathematical Medicine,2006(5):518-520.
[4]Cyr L,Francis K.Measures of clinical agreement for nominal and categorical data:The Kappa coefficient[J].Computer Sin Biology and Medicine,1992,22(4):239-246.
[5]文万青.关于Kappa系数的进一步评价和校正[J].中国卫生统计,1994(6):4-7.Wen Wanqing.Further evaluation and correction of Kappa coefficient[J].Chinese Journal of Health Statistics,1994(6):4-7.
[6]李思泉,张轩雄.基于卷积神经网络的人脸表情识别研究[J].软件导刊,2018,17(1):28-31.Li Siquan,Zhang Xuanxiong.Study on face expression recognition based on convolutional neural network[J].Software Guide,2018,17(1):28-31.
[7]陈海永,徐森,刘坤,等.基于Gabor小波和加权马氏距离的带钢表面缺陷检测[J].电子测量与仪器学报,2016,30(5):786-793.Chen Haiyong,Xu Sen,Liu Kun,et al.Strip defect detection based on Gabor wavelet and weighted Mahalanobis distance[J].Journal of Electronic Measurement and Instrumentation,2016,30(5):786-793.
[8]陈伏兵,陈秀宏,张生亮,等.基于模块2DPCA的人脸识别方法[J].中国图象图形学报,2006(4):580-585.Chen Fubing,Chen Xiuhong,Zhang Shengliang,et al.A human face recognition method based onmodular 2DPCA[J].Journal of Image and Graphics,2006(4):580-585.
[9]颜雪军,赵春霞,袁夏.2DPCA-SIFT:一种有效的局部特征描述方法[J].自动化学报,2014,40(4):675-682.Yan Xuejun,Zhao Chunxia,Yuan Xia.2DPCA-SIFT:An efficient local feature descriptor[J].Acta Automatica Sinica,2014,40(4):675-682.
[10]赵丽美,贾维敏,王标标,等.基于最大相关熵准则的2维主成分分析[J].中国图象图形学报,2015,20(12):1684-1688.Zhao Limei,Jia Weimin,Wang Biaobiao,et al.Two-dimensional principal component analysis based on maximum correntropy criterion[J].Journal of Image and Graphics,2015,20(12):1684-1688.
[11]杨勇,童松,黄淑英.快速离散Curvelet变换域的图像融合[J].中国图象图形学报,2015,20(2):219-228.Yang Yong,Tong Song,Huang Shuying.Image fusion based on fast discrete Curvelet transform[J].Journal of Image and Graphics,2015,20(2):219-228.
[12]Bai Yun,Li Yong,Wang Xiaoxue,et al.Air pollutants concentrations forecasting using back propagation neural network based on wavelet decomposition with meteorological conditions[J].Atmospheric Pollution Research,2016,7(3):1-5.
[13]Krizhevsky A,Sutskever I,Hinton G E.ImageNet classification with deep convolutional neural networks[J].Communications of the ACM,2012,60(6):84-90.
[14]LeCun Y,Boser B,Denker J S,et al.Backpropagation applied to handwritten zip code recognition[J].Neural Computation,1989,1(4):541-551.
[15]Zeiler M D,Fergus R.Visualizing and understanding convolutional networks[C]//Computer Vision-ECCV 2014.[S.l.]:Springer International Publishing,2014:818.
[2]甘荧.自然场景下人脸表情数据集的构建与应用研究[D].南昌:江西师范大学,2016.Gan Ying.Construction and application research of facial expression datasets in wild[D].Nanchang:Jiangxi Normal University,2016.
[3]华琳,阎岩,张建.关于对诊断一致性Kappa系统的探讨[J].数理医药学杂志,2006(5):518-520.Hua Lin,Yan Yan,Zhang Jian.Discussion on diagnostic consistency Kappa system[J].Journal of Mathematical Medicine,2006(5):518-520.
[4]Cyr L,Francis K.Measures of clinical agreement for nominal and categorical data:The Kappa coefficient[J].Computer Sin Biology and Medicine,1992,22(4):239-246.
[5]文万青.关于Kappa系数的进一步评价和校正[J].中国卫生统计,1994(6):4-7.Wen Wanqing.Further evaluation and correction of Kappa coefficient[J].Chinese Journal of Health Statistics,1994(6):4-7.
[6]李思泉,张轩雄.基于卷积神经网络的人脸表情识别研究[J].软件导刊,2018,17(1):28-31.Li Siquan,Zhang Xuanxiong.Study on face expression recognition based on convolutional neural network[J].Software Guide,2018,17(1):28-31.
[7]陈海永,徐森,刘坤,等.基于Gabor小波和加权马氏距离的带钢表面缺陷检测[J].电子测量与仪器学报,2016,30(5):786-793.Chen Haiyong,Xu Sen,Liu Kun,et al.Strip defect detection based on Gabor wavelet and weighted Mahalanobis distance[J].Journal of Electronic Measurement and Instrumentation,2016,30(5):786-793.
[8]陈伏兵,陈秀宏,张生亮,等.基于模块2DPCA的人脸识别方法[J].中国图象图形学报,2006(4):580-585.Chen Fubing,Chen Xiuhong,Zhang Shengliang,et al.A human face recognition method based onmodular 2DPCA[J].Journal of Image and Graphics,2006(4):580-585.
[9]颜雪军,赵春霞,袁夏.2DPCA-SIFT:一种有效的局部特征描述方法[J].自动化学报,2014,40(4):675-682.Yan Xuejun,Zhao Chunxia,Yuan Xia.2DPCA-SIFT:An efficient local feature descriptor[J].Acta Automatica Sinica,2014,40(4):675-682.
[10]赵丽美,贾维敏,王标标,等.基于最大相关熵准则的2维主成分分析[J].中国图象图形学报,2015,20(12):1684-1688.Zhao Limei,Jia Weimin,Wang Biaobiao,et al.Two-dimensional principal component analysis based on maximum correntropy criterion[J].Journal of Image and Graphics,2015,20(12):1684-1688.
[11]杨勇,童松,黄淑英.快速离散Curvelet变换域的图像融合[J].中国图象图形学报,2015,20(2):219-228.Yang Yong,Tong Song,Huang Shuying.Image fusion based on fast discrete Curvelet transform[J].Journal of Image and Graphics,2015,20(2):219-228.
[12]Bai Yun,Li Yong,Wang Xiaoxue,et al.Air pollutants concentrations forecasting using back propagation neural network based on wavelet decomposition with meteorological conditions[J].Atmospheric Pollution Research,2016,7(3):1-5.
[13]Krizhevsky A,Sutskever I,Hinton G E.ImageNet classification with deep convolutional neural networks[J].Communications of the ACM,2012,60(6):84-90.
[14]LeCun Y,Boser B,Denker J S,et al.Backpropagation applied to handwritten zip code recognition[J].Neural Computation,1989,1(4):541-551.
[15]Zeiler M D,Fergus R.Visualizing and understanding convolutional networks[C]//Computer Vision-ECCV 2014.[S.l.]:Springer International Publishing,2014:818.