Pub Date : 2013-04-04DOI: 10.1109/IWBF.2013.6547321
Silvia Gabriel-Sanz, R. Vera-Rodríguez, Pedro Tome, Julian Fierrez
This paper is focused on the assessment of gait recognition on a constrained scenario, where limited information can be extracted from the gait image sequences. In particular we are interested in assessing the performance of gait images when only the lower part of the body is acquired by the camera and just half of a gait cycle is available (SFootBD database). Thus, various state-of-the-art feature approaches have been followed and applied to the data. A comparison with a standard and ideal gait database (USF database) is also carried out using similar experimental protocols. Results show that good recognition performance can be achieved using such limited data information for gait biometric (around 85% of rank 5 identification rate and 8.6% of EER). The comparison with a standard database shows that different feature approaches perform differently for each database, achieving best individual results with MPCA and EGEI methods for the SFootBD and the USF database respectively.
{"title":"Assessment of gait recognition based on the lower part of the human body","authors":"Silvia Gabriel-Sanz, R. Vera-Rodríguez, Pedro Tome, Julian Fierrez","doi":"10.1109/IWBF.2013.6547321","DOIUrl":"https://doi.org/10.1109/IWBF.2013.6547321","url":null,"abstract":"This paper is focused on the assessment of gait recognition on a constrained scenario, where limited information can be extracted from the gait image sequences. In particular we are interested in assessing the performance of gait images when only the lower part of the body is acquired by the camera and just half of a gait cycle is available (SFootBD database). Thus, various state-of-the-art feature approaches have been followed and applied to the data. A comparison with a standard and ideal gait database (USF database) is also carried out using similar experimental protocols. Results show that good recognition performance can be achieved using such limited data information for gait biometric (around 85% of rank 5 identification rate and 8.6% of EER). The comparison with a standard database shows that different feature approaches perform differently for each database, achieving best individual results with MPCA and EGEI methods for the SFootBD and the USF database respectively.","PeriodicalId":412596,"journal":{"name":"2013 International Workshop on Biometrics and Forensics (IWBF)","volume":"106 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132374086","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-04DOI: 10.1109/IWBF.2013.6547307
V. Štruc, N. Pavesic, J. Zganec-Gros
Similarity scores, which form the basis for identity inference in biometric verification systems, typically exhibit statistical variations. These variations are caused by so-called miss-matched conditions, in which the enrollment and probe samples were acquired, and are common to most application domains of biometric verification systems ranging from forensics to smart-home environments. To mitigate these variations, score normalization techniques are usually used. Examples of these techniques include the z-norm, the t-norm or the zt-norm. In this paper we study two-step normalization techniques, such as the zt-norm, and propose a new way of implementing such techniques. Specifically, we propose to implement the first step of the two-step procedure off-line in a non-parametric manner, while the second step is kept unchanged and, hence, performed parametrically. As shown in our face verification experiments, the proposed composite scheme can improve upon the performance of parametric normalization techniques, without an increase in computational complexity, as this is the case with pure non-parametric normalization techniques.
{"title":"Composite score normalization for face verification","authors":"V. Štruc, N. Pavesic, J. Zganec-Gros","doi":"10.1109/IWBF.2013.6547307","DOIUrl":"https://doi.org/10.1109/IWBF.2013.6547307","url":null,"abstract":"Similarity scores, which form the basis for identity inference in biometric verification systems, typically exhibit statistical variations. These variations are caused by so-called miss-matched conditions, in which the enrollment and probe samples were acquired, and are common to most application domains of biometric verification systems ranging from forensics to smart-home environments. To mitigate these variations, score normalization techniques are usually used. Examples of these techniques include the z-norm, the t-norm or the zt-norm. In this paper we study two-step normalization techniques, such as the zt-norm, and propose a new way of implementing such techniques. Specifically, we propose to implement the first step of the two-step procedure off-line in a non-parametric manner, while the second step is kept unchanged and, hence, performed parametrically. As shown in our face verification experiments, the proposed composite scheme can improve upon the performance of parametric normalization techniques, without an increase in computational complexity, as this is the case with pure non-parametric normalization techniques.","PeriodicalId":412596,"journal":{"name":"2013 International Workshop on Biometrics and Forensics (IWBF)","volume":"573 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133879235","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-04DOI: 10.1109/IWBF.2013.6547309
Amani A. Al-Ajlan
Recent researches have shown great attention in biometric systems due to the importance of its applications and its challenges. Researchers proved that the available fingerprint recognition systems are vulnerable to attack at sensor level. Thus, developing an efficient algorithm to protect fingerprint systems from imposer access is urgent. This can be achieved by checking the vitality of a presented fingerprint using hardware or software solutions. This survey focuses on current state of art in the field of fingerprint liveness detection by summarizing the related researches.
{"title":"Survey on fingerprint liveness detection","authors":"Amani A. Al-Ajlan","doi":"10.1109/IWBF.2013.6547309","DOIUrl":"https://doi.org/10.1109/IWBF.2013.6547309","url":null,"abstract":"Recent researches have shown great attention in biometric systems due to the importance of its applications and its challenges. Researchers proved that the available fingerprint recognition systems are vulnerable to attack at sensor level. Thus, developing an efficient algorithm to protect fingerprint systems from imposer access is urgent. This can be achieved by checking the vitality of a presented fingerprint using hardware or software solutions. This survey focuses on current state of art in the field of fingerprint liveness detection by summarizing the related researches.","PeriodicalId":412596,"journal":{"name":"2013 International Workshop on Biometrics and Forensics (IWBF)","volume":"8 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131628305","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-04DOI: 10.1109/IWBF.2013.6547305
Xingjie Wei, Chang-Tsun Li, Yongjian Hu
Face recognition (FR) systems in real environment need to deal with uncontrolled variations in face images such as occlusions and disguise. Most of the current FR algorithms do not consider the fact that occlusions may exist in both reference and query images. In this paper, we summarise three occlusion cases that a realistic FR system should take account of. We present a novel non-parametric classification method to handle the occlusion related problems. Our method represents a face image as a sub-patch sequence which maintains the inherent structure information of the face. Matching is based on the Image-to-Class distance from a query sequence to all reference sequences of an enrolled class. Experimental results on public databases verify the effectiveness and robustness of the proposed method.
{"title":"Robust face recognition with occlusions in both reference and query images","authors":"Xingjie Wei, Chang-Tsun Li, Yongjian Hu","doi":"10.1109/IWBF.2013.6547305","DOIUrl":"https://doi.org/10.1109/IWBF.2013.6547305","url":null,"abstract":"Face recognition (FR) systems in real environment need to deal with uncontrolled variations in face images such as occlusions and disguise. Most of the current FR algorithms do not consider the fact that occlusions may exist in both reference and query images. In this paper, we summarise three occlusion cases that a realistic FR system should take account of. We present a novel non-parametric classification method to handle the occlusion related problems. Our method represents a face image as a sub-patch sequence which maintains the inherent structure information of the face. Matching is based on the Image-to-Class distance from a query sequence to all reference sequences of an enrolled class. Experimental results on public databases verify the effectiveness and robustness of the proposed method.","PeriodicalId":412596,"journal":{"name":"2013 International Workshop on Biometrics and Forensics (IWBF)","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126992353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-04DOI: 10.1109/IWBF.2013.6547308
Ruifang Wang, R. Veldhuis, D. Ramos, L. Spreeuwers, Julian Fierrez, Hai-yun Xu
The spectral minutiae representation has been proposed as a novel method to minutiae-based fingerprint recognition, which can handle minutiae translation and rotation and improve matching speed. As high-resolution palmprint recognition is also mainly based on minutiae sets, we apply spectral minutiae representation to palmprints and implement spectral minutiae based matching. We optimize key parameters for the method by experimental study on the characteristics of spectral minutiae using both fingerprints and palmprints. However, experimental results show that spectral minutiae representation has much worse performance for palmprints than that for fingerprints. EER 15.89% and 14.2% are achieved on the public high-resolution palmprint database THUPALMLAB using location-based spectral minutiae representation (SML) and the complex spectral minutiae representation (SMC) respectively while 5.1% and 3.05% on FVC2002 DB2A fingerprint database. Based on statistical analysis, we find the worse performance for palmprints mainly due to larger non-linear distortion and much larger number of minutiae.
{"title":"On the use of spectral minutiae in high-resolution palmprint recognition","authors":"Ruifang Wang, R. Veldhuis, D. Ramos, L. Spreeuwers, Julian Fierrez, Hai-yun Xu","doi":"10.1109/IWBF.2013.6547308","DOIUrl":"https://doi.org/10.1109/IWBF.2013.6547308","url":null,"abstract":"The spectral minutiae representation has been proposed as a novel method to minutiae-based fingerprint recognition, which can handle minutiae translation and rotation and improve matching speed. As high-resolution palmprint recognition is also mainly based on minutiae sets, we apply spectral minutiae representation to palmprints and implement spectral minutiae based matching. We optimize key parameters for the method by experimental study on the characteristics of spectral minutiae using both fingerprints and palmprints. However, experimental results show that spectral minutiae representation has much worse performance for palmprints than that for fingerprints. EER 15.89% and 14.2% are achieved on the public high-resolution palmprint database THUPALMLAB using location-based spectral minutiae representation (SML) and the complex spectral minutiae representation (SMC) respectively while 5.1% and 3.05% on FVC2002 DB2A fingerprint database. Based on statistical analysis, we find the worse performance for palmprints mainly due to larger non-linear distortion and much larger number of minutiae.","PeriodicalId":412596,"journal":{"name":"2013 International Workshop on Biometrics and Forensics (IWBF)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133813784","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-04DOI: 10.1109/IWBF.2013.6547318
Sanchit Singh, P. Correia, Luís Ducla Soares
This paper presents a further improvement of the previously published technique, PP-RIDER (Partial Palmprint Rotation-Invariant and DEgraded Recognition). PP-RIDER technique combines Modified Phase-Only Correlation (MPOC) and Fourier-Mellin Transform (FMT), and was proposed for recognizing randomly-rotated and degraded partial palmprints with respect to full palmprints registered in a database. In this paper, the influence of the inside-lobe size for the MPOC technique in PP-RIDER is analyzed. In fact, MPOC may produce spurious peaks around the main correlation peak which, when using too small inside-lobe sizes, can affect negatively the similarity scores in genuine comparisons. The proposed improved version of PP-RIDER is tested using the modified THUPALMLAB and PV-TEST-PARTIAL databases. The experimental results show that the proposed modifications allow achieving an improved performance when compared to the baseline technique.
{"title":"Improved rotation-invariant degraded partial palmprint recognition technique","authors":"Sanchit Singh, P. Correia, Luís Ducla Soares","doi":"10.1109/IWBF.2013.6547318","DOIUrl":"https://doi.org/10.1109/IWBF.2013.6547318","url":null,"abstract":"This paper presents a further improvement of the previously published technique, PP-RIDER (Partial Palmprint Rotation-Invariant and DEgraded Recognition). PP-RIDER technique combines Modified Phase-Only Correlation (MPOC) and Fourier-Mellin Transform (FMT), and was proposed for recognizing randomly-rotated and degraded partial palmprints with respect to full palmprints registered in a database. In this paper, the influence of the inside-lobe size for the MPOC technique in PP-RIDER is analyzed. In fact, MPOC may produce spurious peaks around the main correlation peak which, when using too small inside-lobe sizes, can affect negatively the similarity scores in genuine comparisons. The proposed improved version of PP-RIDER is tested using the modified THUPALMLAB and PV-TEST-PARTIAL databases. The experimental results show that the proposed modifications allow achieving an improved performance when compared to the baseline technique.","PeriodicalId":412596,"journal":{"name":"2013 International Workshop on Biometrics and Forensics (IWBF)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115791848","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-04DOI: 10.1109/IWBF.2013.6547324
Jens Fagertun, Tobias Andersen, Thomas F. Hansen, R. Paulsen
We use 3D scans of human faces and cognitive modeling to estimate the “gender strength”. The “gender strength” is a continuous class variable of the gender, superseding the traditional binary class labeling. To visualize some of the visual trends humans use when performing gender classification, we use linear regression. In addition, we use the gender strength to construct a smaller but refined training set, by identifying and removing ill-defined training examples. We use this refined training set to improve the performance of known classification algorithms. Results are presented using a 5-fold cross-validation scheme and also reproduced using an unseen data set.
{"title":"3D gender recognition using cognitive modeling","authors":"Jens Fagertun, Tobias Andersen, Thomas F. Hansen, R. Paulsen","doi":"10.1109/IWBF.2013.6547324","DOIUrl":"https://doi.org/10.1109/IWBF.2013.6547324","url":null,"abstract":"We use 3D scans of human faces and cognitive modeling to estimate the “gender strength”. The “gender strength” is a continuous class variable of the gender, superseding the traditional binary class labeling. To visualize some of the visual trends humans use when performing gender classification, we use linear regression. In addition, we use the gender strength to construct a smaller but refined training set, by identifying and removing ill-defined training examples. We use this refined training set to improve the performance of known classification algorithms. Results are presented using a 5-fold cross-validation scheme and also reproduced using an unseen data set.","PeriodicalId":412596,"journal":{"name":"2013 International Workshop on Biometrics and Forensics (IWBF)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128653503","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-04DOI: 10.1109/IWBF.2013.6547315
J. Sturm, M. Hildebrandt, J. Dittmann, C. Vielhauer
Schwarz's technique for printing amino acid solutions introduces the possibility of printing large quantities of latent fingerprints for crime scene investigation quality assurance. Nevertheless his technique also unintentionally creates the possibility of leaving printed fingerprints at crime scenes. To help identify those false fingerprints, in our paper we extend the printing pipeline, for training investigators and detection methods. Furthermore, we propose subjective and objective evaluation approaches and first tendencies for boundary ranges for objective evaluation metrics. In particular we use digitized real latent fingerprints as printing source (= template) and different contactless sensors (two different chromatic white light sensors, FRT CWL 600, FRT CWL 1mm, and a confocal microscope Keyence VK-X105) for their acquisition. For the examination of the printed fingerprints one subjective and two objective evaluation approaches are introduced as well as a first tendency for boundary ranges of the objective approach. A Canon PIXMA IP 4600 is used for printing and the Keyence VK-X105 acquires the untreated printed fingerprints. Our benchmarking results show that the acquisition sensor Keyence VK-X105 leads to the highest quality of printed fingerprints. In respect to the boundary ranges our suggested first tendency is: correlation value with 20x-objective: Best = [0,...,0.1150], Average = [0.1151,...,0.1258], Worst = [0.1259,...,1]. With 50x-objective: Best = [0,...,0.1299], Average = [0.1300,..., 0.1443], Worst = [0.1444,...,1]. And for the average value with 20x-objective: Best = [0,...,0.0171], Average = [0.0172,...,0.0260], Worst = [0.0261,...,1]. And with 50x-objective: Best = [0,...,0.0299], Average = [0.0300,...,0.0470], Worst = [0.0471,...,1].
{"title":"High quality training materials to detect printed fingerprints: Benchmarking three different aquisition sensors producing printing templates","authors":"J. Sturm, M. Hildebrandt, J. Dittmann, C. Vielhauer","doi":"10.1109/IWBF.2013.6547315","DOIUrl":"https://doi.org/10.1109/IWBF.2013.6547315","url":null,"abstract":"Schwarz's technique for printing amino acid solutions introduces the possibility of printing large quantities of latent fingerprints for crime scene investigation quality assurance. Nevertheless his technique also unintentionally creates the possibility of leaving printed fingerprints at crime scenes. To help identify those false fingerprints, in our paper we extend the printing pipeline, for training investigators and detection methods. Furthermore, we propose subjective and objective evaluation approaches and first tendencies for boundary ranges for objective evaluation metrics. In particular we use digitized real latent fingerprints as printing source (= template) and different contactless sensors (two different chromatic white light sensors, FRT CWL 600, FRT CWL 1mm, and a confocal microscope Keyence VK-X105) for their acquisition. For the examination of the printed fingerprints one subjective and two objective evaluation approaches are introduced as well as a first tendency for boundary ranges of the objective approach. A Canon PIXMA IP 4600 is used for printing and the Keyence VK-X105 acquires the untreated printed fingerprints. Our benchmarking results show that the acquisition sensor Keyence VK-X105 leads to the highest quality of printed fingerprints. In respect to the boundary ranges our suggested first tendency is: correlation value with 20x-objective: Best = [0,...,0.1150], Average = [0.1151,...,0.1258], Worst = [0.1259,...,1]. With 50x-objective: Best = [0,...,0.1299], Average = [0.1300,..., 0.1443], Worst = [0.1444,...,1]. And for the average value with 20x-objective: Best = [0,...,0.0171], Average = [0.0172,...,0.0260], Worst = [0.0261,...,1]. And with 50x-objective: Best = [0,...,0.0299], Average = [0.0300,...,0.0470], Worst = [0.0471,...,1].","PeriodicalId":412596,"journal":{"name":"2013 International Workshop on Biometrics and Forensics (IWBF)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125408312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-04DOI: 10.1109/IWBF.2013.6547314
Jesus Valera, N. García
A method is proposed for unambiguously determining the interpolation rate in uncompressed resized color images over a broad range of possible rates, thus overcoming the uncertainty exposed in previous papers. Information from the three RGB channels is combined to improve performance. Experimental results show high reliability in the posteriori probability of correct classification, yielding an average value of 81.7% and 90.3% in the case of images with a CFA type sensor. The proposed method is also used for the automatic detection of CFA traces with good results.
{"title":"Unambiguous interpolation rate estimation in uncompressed resized color digital images","authors":"Jesus Valera, N. García","doi":"10.1109/IWBF.2013.6547314","DOIUrl":"https://doi.org/10.1109/IWBF.2013.6547314","url":null,"abstract":"A method is proposed for unambiguously determining the interpolation rate in uncompressed resized color images over a broad range of possible rates, thus overcoming the uncertainty exposed in previous papers. Information from the three RGB channels is combined to improve performance. Experimental results show high reliability in the posteriori probability of correct classification, yielding an average value of 81.7% and 90.3% in the case of images with a CFA type sensor. The proposed method is also used for the automatic detection of CFA traces with good results.","PeriodicalId":412596,"journal":{"name":"2013 International Workshop on Biometrics and Forensics (IWBF)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134094432","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-04DOI: 10.1109/IWBF.2013.6547325
S. Harder, L. Clemmensen, Anders L. Dahl, J. D. Andersen, P. Johansen, Susanne R. Christoffersen, N. Morling, C. Børsting, R. Paulsen
The presented work concerns prediction of complex human phenotypes from genotypes. We were interested in correlating iris color and texture with DNA. Our data consist of 212 eye images along with DNA: 32 single-nucleotide polymorphisms (SNPs). We used two types of biometrics to describe the eye images: One for iris color and one for iris texture. Both biometrics were high dimensional and a sparse principle component analysis (SPCA) reduced the dimensions and resulted in a representation of data with good interpretability. The correlations between the sparse principal components (SPCs) and the 32 SNPs were found using a canonical correlation analysis (CCA). The result was a single significant canonical correlation (CC) for both biometrics. Each CC comprised two correlated canonical variables, consisting of a linear combination of SPCs and a linear combination of SNPs, respectively. The significant canonical variables for color and texture were primarily explained by the first SPC (SPC1). Therefore, we made a visual inspection of the first SPCs. The color based SPC1 explained a blue to brown variation in iris color and the texture based SPC1 gave a general explanation of iris texture. The SNPs (rs12896399, rs3733542, rs6475555, rs12913832) and (rs12896399, rs3733542, rs12913832) had the highest correlation to the canonical variable for color and texture, respectively. Three of the most contributing SNPs were the same for both biometrics, revealing a covariance between iris color and texture.
介绍的工作涉及从基因型预测复杂的人类表型。我们感兴趣的是将虹膜的颜色和纹理与 DNA 联系起来。我们的数据包括 212 幅眼睛图像和 DNA：32 个单核苷酸多态性 (SNP)。我们使用两种生物统计学方法来描述眼睛图像：一种是虹膜颜色，另一种是虹膜纹理。这两种生物统计量的维度都很高，稀疏原理成分分析（SPCA）降低了维度，使数据表示具有良好的可解释性。稀疏主成分 (SPC) 与 32 个 SNP 之间的相关性是通过典型相关分析 (CCA) 发现的。结果发现，这两种生物统计学指标之间存在单个显著的典型相关性（CC）。每个 CC 包括两个相关的典型变量，分别由 SPC 的线性组合和 SNP 的线性组合组成。颜色和纹理的重要典型变量主要由第一个 SPC（SPC1）解释。因此，我们对第一个 SPC 进行了目测。基于颜色的 SPC1 解释了虹膜颜色从蓝色到棕色的变化，而基于纹理的 SPC1 则对虹膜纹理做出了一般性解释。SNP（rs12896399、rs3733542、rs6475555、rs12913832）和（rs12896399、rs3733542、rs12913832）分别与颜色和纹理的典型变量具有最高的相关性。两个生物统计学中贡献最大的 SNPs 有三个是相同的，这揭示了虹膜颜色和纹理之间的协方差。
{"title":"Correlation of iris biometrics and DNA","authors":"S. Harder, L. Clemmensen, Anders L. Dahl, J. D. Andersen, P. Johansen, Susanne R. Christoffersen, N. Morling, C. Børsting, R. Paulsen","doi":"10.1109/IWBF.2013.6547325","DOIUrl":"https://doi.org/10.1109/IWBF.2013.6547325","url":null,"abstract":"The presented work concerns prediction of complex human phenotypes from genotypes. We were interested in correlating iris color and texture with DNA. Our data consist of 212 eye images along with DNA: 32 single-nucleotide polymorphisms (SNPs). We used two types of biometrics to describe the eye images: One for iris color and one for iris texture. Both biometrics were high dimensional and a sparse principle component analysis (SPCA) reduced the dimensions and resulted in a representation of data with good interpretability. The correlations between the sparse principal components (SPCs) and the 32 SNPs were found using a canonical correlation analysis (CCA). The result was a single significant canonical correlation (CC) for both biometrics. Each CC comprised two correlated canonical variables, consisting of a linear combination of SPCs and a linear combination of SNPs, respectively. The significant canonical variables for color and texture were primarily explained by the first SPC (SPC1). Therefore, we made a visual inspection of the first SPCs. The color based SPC1 explained a blue to brown variation in iris color and the texture based SPC1 gave a general explanation of iris texture. The SNPs (rs12896399, rs3733542, rs6475555, rs12913832) and (rs12896399, rs3733542, rs12913832) had the highest correlation to the canonical variable for color and texture, respectively. Three of the most contributing SNPs were the same for both biometrics, revealing a covariance between iris color and texture.","PeriodicalId":412596,"journal":{"name":"2013 International Workshop on Biometrics and Forensics (IWBF)","volume":"79 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130983267","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：