融合皮肤色素分布特性及最佳尺度的毛孔检测及评价
|
摘要
面部毛孔特征是人脸识别、皮肤检测的重要指标之一.为了有效地消除其他皮肤特征对毛孔检测的影响,提出一种融合皮肤色素分布特性及最佳尺度的高精度毛孔检测算法.针对皮肤特征在不同色素层上的色素分布特性相异,首先根据皮肤特征显著性差异和K-means聚类为SURF和SIFT算法设置合理阈值检测不同色素层图像上的皮肤特征;然后引入欧氏距离描述不同色素层检测点的位置信息相似性,利用最佳尺度作为阈值有效地筛除非毛孔特征干扰项.在毛孔检测的基础上,利用SIFT算法中的最佳尺度构建了皮肤毛孔粗糙度评价指标.选取Bosphorus人脸库中的正脸图像进行毛孔检测及评价的对比实验,结果表明该算法提高了毛孔检测的准确度,构建的评价指标稳定有效.
Facial pore feature is one of important indicators for face recognition and skin defects detection. In order to eliminate the effect of other skin features from facial pore detection processing effectively, we proposed a new facial pore detection algorithm that coalesces the characteristics of skin pigment distribution and optimal scale. Considering the dissimilarity of skin pigment distribution on different pigment layers, reasonable thresholds were set for SURF and SIFT algorithms to detect the skin features on different pigment layers, which was based on the significant difference of skin features and K-means clustering method. Then, the Euclidean distance was calculated to describe the positions similarity of the same detected points on different layers. Last, the optimal scales were set as thresholds to screen off the interferences of skin features except pore. On this basis, an evaluation criterion of skin pores roughness was also established by using the optimal scales obtained during SIFT process. We selected positive face images from Bosphorus dataset for comparative experiments about pore detection and evaluation. The experiment results show that the proposed algorithm improves accuracy of pore detection,and the evaluation criterion is stable and effective.
Facial pore feature is one of important indicators for face recognition and skin defects detection. In order to eliminate the effect of other skin features from facial pore detection processing effectively, we proposed a new facial pore detection algorithm that coalesces the characteristics of skin pigment distribution and optimal scale. Considering the dissimilarity of skin pigment distribution on different pigment layers, reasonable thresholds were set for SURF and SIFT algorithms to detect the skin features on different pigment layers, which was based on the significant difference of skin features and K-means clustering method. Then, the Euclidean distance was calculated to describe the positions similarity of the same detected points on different layers. Last, the optimal scales were set as thresholds to screen off the interferences of skin features except pore. On this basis, an evaluation criterion of skin pores roughness was also established by using the optimal scales obtained during SIFT process. We selected positive face images from Bosphorus dataset for comparative experiments about pore detection and evaluation. The experiment results show that the proposed algorithm improves accuracy of pore detection,and the evaluation criterion is stable and effective.
引文
[1]Sun J Y,Kim S W,Lee S H,et al.Automatic facial pore analysis system using multi‐scale pore detection[J].Skin Research&Technology,2016,23(3):354-362
[2]Wang X D,Liang Y W,Zeng X X,et al.Deeply learned pore-scale facial features[C]//Proceedings of Chinese Conference on Biometric Recognition.Heidelberg:Springer,2017:135-144
[3]Mc Grath J A,Eady R A J,Pope F M,et al.Rook’s textbook of dermatology[M].7th eds.Oxford:Blackwell Press,2004:45-128
[4]Malathi S,Meena C.Improved partial fingerprint matching based on score level fusion using pore and SIFT features[C]//Proceedings of International Conference on Process Automation,Control and Computing.Los Alamitos:IEEE Computer Society Press,2011:1-4
[5]Lowe D G.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91-110
[6]Song S L,Ohnuma K,Liu Z,et al.Novel biometrics based on nose pore recognition[J].Optical Engineering,2009,48(5):Article No.057204
[7]Song Shangling,Kazuhiko Ohnuma,Mei Liangmo,et al.Biometric recognition based on nose pore feature[J].Pattern Recognition and Artificial Intelligence,2009,22(6):919-923(in Chinese)(宋尚玲,大沼一彦,梅良模,等.鼻部毛囊的生物特征识别[J].模式识别与人工智能,2009,22(6):919-923)
[8]Li D,Lam K M.Design and learn distinctive features from pore-scale facial keypoints[J].Pattern Recognition,2015,48(3):732-745
[9]Tsumura N,Ojima N,Sato K,et al.Imaged-based skin color and texture analysis/synthesis by extracting hemoglobin and melanin information in the skin[J].ACM Transactions on Graphics,2003,22(3):770-779
[10]Tsumura N,Haneishi H,Miyake Y.Independent-component analysis of skin color image[J].Journal of the Optical Society of America,1999,16(9):2169-2176
[11]Villegas-Quezada C,Climent J,Rodngo Villegas C.Using skin melanin layer for facial pore identification in RGB digital images[J].International Journal of Emerging Technology and Advanced Engineering,2014,4(8):335-342
[12]Villegas-Quezada C,Climent J.Facial pores in melanin layer and genetic algorithms for human face characterization[J].Frontiers in Pathology and Genetics,2014,2:12-21
[13]Liu Z,Josiane Z.Melanin and hemoglobin identification for skin disease analysys[C]//Proceedings of the 2nd IAPR Asian Conference on Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2014:145-149
[14]Jolivot R,Benezeth Y,Marzani F.Skin parameter map retrieval from a dedicated multispectral imaging system applied to dermatology/cosmetology[J].International Journal of Biomedical Imaging,2013:Article No.978289
[15]Kharkar V P,Ratnaparkhe V P.Hemoglobin estimation methods:a review of clinical,sensor and image processing methods[J].International Journal of Engineering Research&Technology,2013,2(1):Article ID IJERTV2IS1364
[16]Francois G,Maudet A,McDaniel D,et al.Quantification of facial pores using image analysis[J].Cosmetic Dermatology,2009,22(9):457-463
[17]Flament F,Francois G,Qiu H X,et al.Facial skin pores:a multiethnic study[J].Clinical,Cosmetic and Investigational Dermatology,2015,8:85-93
[18]Yao N,Zeng Q,Wang Q,et al.Evaluating photographic scales of facial pores and diagnostic agreement of tests using latent class models[J].Journal of Cutaneous and Laser Therapy,2017,19(1):64-67
[19]Shaiek A,Flament F,Fran?ois G,et al.A new tool to quantify the geometrical characteristics of facial skin pores.Changes with age and a making-up procedure in Caucasian women[J].Skin Research&Technology,2017,23(2):249-257
[20]Joswig J O,Lorenz T.Detecting and quantifying geometric features in large series of cluster structures[J].Zeitschrift für Physikalische Chemie,2016,230(5-7):1057-1066
[21]Jang S I,Kim E J,Lee H K.A method of evaluating facial pores using optical 2D images and analysis of age-dependent changes in facial pores in Koreans[J].Skin Research&Technology,2018,24(2):304-308
[22]Jung H J,Ahn J Y,Lee J I,et al.Analysis of the number of enlarged pores according to site,age,and sex[J].Skin Research&Technology,2018,24(3):367-370
[23]Bay H,Ess A,Tuytelaars T,et al.Speeded-up robust features(SURF)[J].Computer Vision and Image Understanding,2008,110(3):346-359
[24]Hiraoka M,Filrbank M,Essenpreis M,et al.A Monte Carlo investigation of optical pathlength in inhomogeneous tissue and its application to near-infrared spectroscopy[J].Physics in Medicine and Biology,1993,38(12):1859-1876
[25]Xu Shuchang,Zhang Sanyuan,Zhang Yin,et al.Robust algorithm for extracting skin pigment concentration from color image[J].Journal of Zhejiang University:Engineering Science,2011,45(2):253-258(in Chinese)(徐舒畅,张三元,张引,等.基于彩色图像的皮肤色素浓度提取算法[J].浙江大学学报:工学版,2011,45(2):253-258)
[26]Tamura H,Mori S,Yamawaki T.Textural features corresponding to visual perception[J].IEEE Transactions on Systems,Man and Cybernetics,1978,8(6):460-473
[27]Savran A,Alyüz N,Dibeklio?lu H,et al.Biometrics and identity management[M].Heidelberg:Springer,2008:47-56
[2]Wang X D,Liang Y W,Zeng X X,et al.Deeply learned pore-scale facial features[C]//Proceedings of Chinese Conference on Biometric Recognition.Heidelberg:Springer,2017:135-144
[3]Mc Grath J A,Eady R A J,Pope F M,et al.Rook’s textbook of dermatology[M].7th eds.Oxford:Blackwell Press,2004:45-128
[4]Malathi S,Meena C.Improved partial fingerprint matching based on score level fusion using pore and SIFT features[C]//Proceedings of International Conference on Process Automation,Control and Computing.Los Alamitos:IEEE Computer Society Press,2011:1-4
[5]Lowe D G.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91-110
[6]Song S L,Ohnuma K,Liu Z,et al.Novel biometrics based on nose pore recognition[J].Optical Engineering,2009,48(5):Article No.057204
[7]Song Shangling,Kazuhiko Ohnuma,Mei Liangmo,et al.Biometric recognition based on nose pore feature[J].Pattern Recognition and Artificial Intelligence,2009,22(6):919-923(in Chinese)(宋尚玲,大沼一彦,梅良模,等.鼻部毛囊的生物特征识别[J].模式识别与人工智能,2009,22(6):919-923)
[8]Li D,Lam K M.Design and learn distinctive features from pore-scale facial keypoints[J].Pattern Recognition,2015,48(3):732-745
[9]Tsumura N,Ojima N,Sato K,et al.Imaged-based skin color and texture analysis/synthesis by extracting hemoglobin and melanin information in the skin[J].ACM Transactions on Graphics,2003,22(3):770-779
[10]Tsumura N,Haneishi H,Miyake Y.Independent-component analysis of skin color image[J].Journal of the Optical Society of America,1999,16(9):2169-2176
[11]Villegas-Quezada C,Climent J,Rodngo Villegas C.Using skin melanin layer for facial pore identification in RGB digital images[J].International Journal of Emerging Technology and Advanced Engineering,2014,4(8):335-342
[12]Villegas-Quezada C,Climent J.Facial pores in melanin layer and genetic algorithms for human face characterization[J].Frontiers in Pathology and Genetics,2014,2:12-21
[13]Liu Z,Josiane Z.Melanin and hemoglobin identification for skin disease analysys[C]//Proceedings of the 2nd IAPR Asian Conference on Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2014:145-149
[14]Jolivot R,Benezeth Y,Marzani F.Skin parameter map retrieval from a dedicated multispectral imaging system applied to dermatology/cosmetology[J].International Journal of Biomedical Imaging,2013:Article No.978289
[15]Kharkar V P,Ratnaparkhe V P.Hemoglobin estimation methods:a review of clinical,sensor and image processing methods[J].International Journal of Engineering Research&Technology,2013,2(1):Article ID IJERTV2IS1364
[16]Francois G,Maudet A,McDaniel D,et al.Quantification of facial pores using image analysis[J].Cosmetic Dermatology,2009,22(9):457-463
[17]Flament F,Francois G,Qiu H X,et al.Facial skin pores:a multiethnic study[J].Clinical,Cosmetic and Investigational Dermatology,2015,8:85-93
[18]Yao N,Zeng Q,Wang Q,et al.Evaluating photographic scales of facial pores and diagnostic agreement of tests using latent class models[J].Journal of Cutaneous and Laser Therapy,2017,19(1):64-67
[19]Shaiek A,Flament F,Fran?ois G,et al.A new tool to quantify the geometrical characteristics of facial skin pores.Changes with age and a making-up procedure in Caucasian women[J].Skin Research&Technology,2017,23(2):249-257
[20]Joswig J O,Lorenz T.Detecting and quantifying geometric features in large series of cluster structures[J].Zeitschrift für Physikalische Chemie,2016,230(5-7):1057-1066
[21]Jang S I,Kim E J,Lee H K.A method of evaluating facial pores using optical 2D images and analysis of age-dependent changes in facial pores in Koreans[J].Skin Research&Technology,2018,24(2):304-308
[22]Jung H J,Ahn J Y,Lee J I,et al.Analysis of the number of enlarged pores according to site,age,and sex[J].Skin Research&Technology,2018,24(3):367-370
[23]Bay H,Ess A,Tuytelaars T,et al.Speeded-up robust features(SURF)[J].Computer Vision and Image Understanding,2008,110(3):346-359
[24]Hiraoka M,Filrbank M,Essenpreis M,et al.A Monte Carlo investigation of optical pathlength in inhomogeneous tissue and its application to near-infrared spectroscopy[J].Physics in Medicine and Biology,1993,38(12):1859-1876
[25]Xu Shuchang,Zhang Sanyuan,Zhang Yin,et al.Robust algorithm for extracting skin pigment concentration from color image[J].Journal of Zhejiang University:Engineering Science,2011,45(2):253-258(in Chinese)(徐舒畅,张三元,张引,等.基于彩色图像的皮肤色素浓度提取算法[J].浙江大学学报:工学版,2011,45(2):253-258)
[26]Tamura H,Mori S,Yamawaki T.Textural features corresponding to visual perception[J].IEEE Transactions on Systems,Man and Cybernetics,1978,8(6):460-473
[27]Savran A,Alyüz N,Dibeklio?lu H,et al.Biometrics and identity management[M].Heidelberg:Springer,2008:47-56