The need for a large sample size grows exponentially with the dimensionality of the feature space ("curse of dimensionality"), which increases the labor cost during the training procedure and severely restricts the number of the practical applications. While feature selection methods can often alleviate the problems associated with the curse of dimensionality, complex large scale pattern recognition problems may not be amenable to features selection approach due to large intrinsic dimensionality. In such situations, the only effective solution to conquer the complications of the high-dimensional functions is to incorporate knowledge about the data that is correct. How to incorporate the domain knowledge with the specific machine learning system has been widely studied in the pattern classification field. In this paper, we will explore a novel method to synthesize a larger, valid training sample data set based on a smaller set of the key samples that are collected by a model based sampling theory that incorporates the domain knowledge of the computer vision. In addition to reducing the training sample size in the learning procedure, our emphasis is on providing practical advice on how to incorporate domain knowledge to design and simplify a vision based pattern classification model.
{"title":"A model-based sampling and sample synthesis method for auto identification in computer vision","authors":"Nanfei Sun, N. Haas, J. Connell, Sharath Pankanti","doi":"10.1109/AUTOID.2005.5","DOIUrl":"https://doi.org/10.1109/AUTOID.2005.5","url":null,"abstract":"The need for a large sample size grows exponentially with the dimensionality of the feature space (\"curse of dimensionality\"), which increases the labor cost during the training procedure and severely restricts the number of the practical applications. While feature selection methods can often alleviate the problems associated with the curse of dimensionality, complex large scale pattern recognition problems may not be amenable to features selection approach due to large intrinsic dimensionality. In such situations, the only effective solution to conquer the complications of the high-dimensional functions is to incorporate knowledge about the data that is correct. How to incorporate the domain knowledge with the specific machine learning system has been widely studied in the pattern classification field. In this paper, we will explore a novel method to synthesize a larger, valid training sample data set based on a smaller set of the key samples that are collected by a model based sampling theory that incorporates the domain knowledge of the computer vision. In addition to reducing the training sample size in the learning procedure, our emphasis is on providing practical advice on how to incorporate domain knowledge to design and simplify a vision based pattern classification model.","PeriodicalId":206458,"journal":{"name":"Fourth IEEE Workshop on Automatic Identification Advanced Technologies (AutoID'05)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116678581","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}
In face recognition, the goal is to assign a class label for a test image of a subject from N classes in the database, when binary classifiers are used, the commonly used method is the one-per-class (OPC) i.e., one classifier per subject. A drawback of the OPC method is that when the number of classes is large, it takes very long time to make a classification decision. In place of the computationally-demanding OPC method, we propose a new feature extraction method "face class code" (FCC) based on binary classifiers. For example, correlation filters and support vector machines can be used to generate feature vectors to deal with large number of classes. The FCC method encodes each class label into a binary string, and we design classifiers to discriminate '1' or '0' for each bit in the sequence, to determine the class label. Thus, we will need as few as [log/sub 2/(N)] binary classifiers to achieve an N-class recognition problem. This binary coding framework also opens the whole world of error control codes (ECC), which can be used to improve the recognition performance. The proposed method is verified through experiments on the PIE database and the AR database.
{"title":"Face class code based feature extraction for face recognition","authors":"C. Xie, B. Kumar","doi":"10.1109/AUTOID.2005.22","DOIUrl":"https://doi.org/10.1109/AUTOID.2005.22","url":null,"abstract":"In face recognition, the goal is to assign a class label for a test image of a subject from N classes in the database, when binary classifiers are used, the commonly used method is the one-per-class (OPC) i.e., one classifier per subject. A drawback of the OPC method is that when the number of classes is large, it takes very long time to make a classification decision. In place of the computationally-demanding OPC method, we propose a new feature extraction method \"face class code\" (FCC) based on binary classifiers. For example, correlation filters and support vector machines can be used to generate feature vectors to deal with large number of classes. The FCC method encodes each class label into a binary string, and we design classifiers to discriminate '1' or '0' for each bit in the sequence, to determine the class label. Thus, we will need as few as [log/sub 2/(N)] binary classifiers to achieve an N-class recognition problem. This binary coding framework also opens the whole world of error control codes (ECC), which can be used to improve the recognition performance. The proposed method is verified through experiments on the PIE database and the AR database.","PeriodicalId":206458,"journal":{"name":"Fourth IEEE Workshop on Automatic Identification Advanced Technologies (AutoID'05)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130903368","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}
Face recognition is a technically difficult task that forms part of an ever-growing number of applications. The challenge becomes more complex by requirement of robustness against variation in facial expressions, pose and lighting condition. In this paper, we present a wavelet-based face recognition scheme and report on its performance on 2 databases. We shall also demonstrate that it is more robust against varying facial expressions than a known scheme that has been developed specifically for that purpose. We shall also report on the positive effect of a simple procedure to reduce the effect of variation in illumination level on accuracy, in contrast to histogram equalization.
{"title":"Face recognition in the presence of expression and/or illumination variation","authors":"H. Sellahewa, S. Jassim","doi":"10.1109/AUTOID.2005.23","DOIUrl":"https://doi.org/10.1109/AUTOID.2005.23","url":null,"abstract":"Face recognition is a technically difficult task that forms part of an ever-growing number of applications. The challenge becomes more complex by requirement of robustness against variation in facial expressions, pose and lighting condition. In this paper, we present a wavelet-based face recognition scheme and report on its performance on 2 databases. We shall also demonstrate that it is more robust against varying facial expressions than a known scheme that has been developed specifically for that purpose. We shall also report on the positive effect of a simple procedure to reduce the effect of variation in illumination level on accuracy, in contrast to histogram equalization.","PeriodicalId":206458,"journal":{"name":"Fourth IEEE Workshop on Automatic Identification Advanced Technologies (AutoID'05)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127923362","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}
In this article, we present three methods for the keystroke dynamic authentication problem. We use in the first method, the average and the standard deviation, in the second the rhythm of striking and in the third, a comparison of the times order. After having presented these methods, we propose to realize a fusion of them. The results obtained indicate good performance of each method alone, as well as a significant improvement of performance with fusion, from 3.43% of EER for the best method alone down-to 1.8% with fusion.
{"title":"Fusion of methods for keystroke dynamic authentication","authors":"Sylvain Hocquet, Jean-Yves Ramel, H. Cardot","doi":"10.1109/AUTOID.2005.30","DOIUrl":"https://doi.org/10.1109/AUTOID.2005.30","url":null,"abstract":"In this article, we present three methods for the keystroke dynamic authentication problem. We use in the first method, the average and the standard deviation, in the second the rhythm of striking and in the third, a comparison of the times order. After having presented these methods, we propose to realize a fusion of them. The results obtained indicate good performance of each method alone, as well as a significant improvement of performance with fusion, from 3.43% of EER for the best method alone down-to 1.8% with fusion.","PeriodicalId":206458,"journal":{"name":"Fourth IEEE Workshop on Automatic Identification Advanced Technologies (AutoID'05)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126801164","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}
Facial recognition/verification R. Chellappa et al., (1995), is a continuing and growing area of research in the field of biometrics. One of the first approaches to this challenge was principal component analysis (PCA) [M. A. Turk et al., (1991), T. Chen et al., (2002)]. Typically PCA is performed in the original spatial domain. However, PCA has a high sensitivity to illumination effects in the original spatial domain. We propose that by using wavelet packet decomposition M. Vetterli et al., (1995), to create localized space-frequency subspaces of the original data, we can perform PCA in these subspaces which can generalize better across illumination variations. We report results on the CMU PIE database T. Sim et al., (2003), by comparing reconstruction error in the original spatial domain to that of the reconstruction error in the spatial subspaces (keeping same number of eigenvectors). It is seen that the total reconstruction error of the space-frequency subspaces is smaller than that of the original space and the automatically pruned wavelet packet PCA produced better face recognition performance across illumination.
面部识别/验证R. Chellappa et al.，(1995)，是生物识别领域一个持续发展的研究领域。应对这一挑战的第一个方法是主成分分析(PCA) [M]。A. Turk et al.， (1991)， T. Chen et al.，(2002)。典型的PCA是在原始空间域中进行的。然而，PCA对原始空间域内的光照效果具有较高的敏感性。我们提出，通过使用小波包分解M. Vetterli等人，(1995)来创建原始数据的局部空间频率子空间，我们可以在这些子空间中执行PCA，从而可以更好地泛化光照变化。我们报告了CMU PIE数据库T. Sim等人(2003)的结果，通过比较原始空间域中的重构误差与空间子空间中的重构误差(保持相同数量的特征向量)。可以看出，空间-频率子空间的总重构误差小于原始空间的总重构误差，并且自动修剪小波包PCA在不同光照下具有更好的人脸识别性能。
{"title":"PCA vs. automatically pruned wavelet-packet PCA for illumination tolerant face recognition","authors":"Ramamurthy Bhagavatula, M. Savvides","doi":"10.1109/AUTOID.2005.38","DOIUrl":"https://doi.org/10.1109/AUTOID.2005.38","url":null,"abstract":"Facial recognition/verification R. Chellappa et al., (1995), is a continuing and growing area of research in the field of biometrics. One of the first approaches to this challenge was principal component analysis (PCA) [M. A. Turk et al., (1991), T. Chen et al., (2002)]. Typically PCA is performed in the original spatial domain. However, PCA has a high sensitivity to illumination effects in the original spatial domain. We propose that by using wavelet packet decomposition M. Vetterli et al., (1995), to create localized space-frequency subspaces of the original data, we can perform PCA in these subspaces which can generalize better across illumination variations. We report results on the CMU PIE database T. Sim et al., (2003), by comparing reconstruction error in the original spatial domain to that of the reconstruction error in the spatial subspaces (keeping same number of eigenvectors). It is seen that the total reconstruction error of the space-frequency subspaces is smaller than that of the original space and the automatically pruned wavelet packet PCA produced better face recognition performance across illumination.","PeriodicalId":206458,"journal":{"name":"Fourth IEEE Workshop on Automatic Identification Advanced Technologies (AutoID'05)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133414516","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}
T. Kevenaar, G. Schrijen, M. V. D. Veen, A. Akkermans, F. Zuo
This paper considers generating binary feature vectors from biometric face data such that their privacy can be protected using recently introduced helper data systems. We explain how the binary feature vectors can be derived and investigate their statistical properties. Experimental results for a subset of the FERET and Caltech databases show that there is only a slight degradation in classification results when using the binary rather than the real-valued feature vectors. Finally, the scheme to extract the binary vectors is combined with a helper data scheme leading to renewable and privacy preserving facial templates with acceptable classification results provided that the within-class variation is not too large.
{"title":"Face recognition with renewable and privacy preserving binary templates","authors":"T. Kevenaar, G. Schrijen, M. V. D. Veen, A. Akkermans, F. Zuo","doi":"10.1109/AUTOID.2005.24","DOIUrl":"https://doi.org/10.1109/AUTOID.2005.24","url":null,"abstract":"This paper considers generating binary feature vectors from biometric face data such that their privacy can be protected using recently introduced helper data systems. We explain how the binary feature vectors can be derived and investigate their statistical properties. Experimental results for a subset of the FERET and Caltech databases show that there is only a slight degradation in classification results when using the binary rather than the real-valued feature vectors. Finally, the scheme to extract the binary vectors is combined with a helper data scheme leading to renewable and privacy preserving facial templates with acceptable classification results provided that the within-class variation is not too large.","PeriodicalId":206458,"journal":{"name":"Fourth IEEE Workshop on Automatic Identification Advanced Technologies (AutoID'05)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126118659","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}
We propose and analyze the effects of attaching more than one RFID tag to each object. We define different types of multi-tag systems and examine their benefits, both analytically and empirically. We also analyze how multi-tags affect some existing tag singulation algorithms. We show how multi-tags can serve as security enhancers, and propose several new promising applications of multi-tags, such as preventing illegal deforestation.
{"title":"Multi-tag radio frequency identification systems","authors":"L. Bolotnyy, G. Robins","doi":"10.1109/AUTOID.2005.36","DOIUrl":"https://doi.org/10.1109/AUTOID.2005.36","url":null,"abstract":"We propose and analyze the effects of attaching more than one RFID tag to each object. We define different types of multi-tag systems and examine their benefits, both analytically and empirically. We also analyze how multi-tags affect some existing tag singulation algorithms. We show how multi-tags can serve as security enhancers, and propose several new promising applications of multi-tags, such as preventing illegal deforestation.","PeriodicalId":206458,"journal":{"name":"Fourth IEEE Workshop on Automatic Identification Advanced Technologies (AutoID'05)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127359739","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}
This work presents a complete control access system based on the hand geometry, a hardware key and a vital sign detector. The circuitry reads the hardware key and the heartbeat in order to confirm the identity of the author given by the analysis of the hand image. The presented system distinguish itself from other similar biometric systems mainly because of the feature extraction process which is based on the analysis of the curvature profile of the image, making the system invariant to the rotation and translation of the hand. This makes unnecessary the use of any kind of restriction devices such as pins or pegs to position the hand. FAR rates as low as 0.8% were obtained by the use of simple weighted geometric features on a database of more than 360 hand images.
{"title":"Hand geometry: a new approach for feature extraction","authors":"Guilherme Boreki, A. Zimmer","doi":"10.1109/AUTOID.2005.33","DOIUrl":"https://doi.org/10.1109/AUTOID.2005.33","url":null,"abstract":"This work presents a complete control access system based on the hand geometry, a hardware key and a vital sign detector. The circuitry reads the hardware key and the heartbeat in order to confirm the identity of the author given by the analysis of the hand image. The presented system distinguish itself from other similar biometric systems mainly because of the feature extraction process which is based on the analysis of the curvature profile of the image, making the system invariant to the rotation and translation of the hand. This makes unnecessary the use of any kind of restriction devices such as pins or pegs to position the hand. FAR rates as low as 0.8% were obtained by the use of simple weighted geometric features on a database of more than 360 hand images.","PeriodicalId":206458,"journal":{"name":"Fourth IEEE Workshop on Automatic Identification Advanced Technologies (AutoID'05)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127878082","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}
Dongjae Lee, Wonchurl Jang, Doeksoo Park, Sung-Jae Kim, Jaihie Kim
This paper proposes a real-time image selection algorithm for fingerprint recognition system, which uses the embedded camera of the mobile device. In general, auto-focusing algorithms of the camera system use the gradient measures to estimate high-frequency components of an image. In mobile device, images are greatly affected by environmental light sources. Therefore, obtained image might not qualify for the fingerprint recognition system, even when the image is focused. Consequently, image should be investigated whether it is usable or not. Variance-modified-Laplacian of Gaussian (VMLOG) and image quality index (QI) proposed in this paper solve such problem. VMLOG considers high-frequency component and repeatable patterns of ridges. Experimental results shows that the processing time of the proposed algorithm is enough fast to be adapted in real time system as like mobile device and the proposed algorithm selects exactly a recognizable image.
{"title":"A real-time image selection algorithm: fingerprint recognition using mobile devices with embedded camera","authors":"Dongjae Lee, Wonchurl Jang, Doeksoo Park, Sung-Jae Kim, Jaihie Kim","doi":"10.1109/AUTOID.2005.6","DOIUrl":"https://doi.org/10.1109/AUTOID.2005.6","url":null,"abstract":"This paper proposes a real-time image selection algorithm for fingerprint recognition system, which uses the embedded camera of the mobile device. In general, auto-focusing algorithms of the camera system use the gradient measures to estimate high-frequency components of an image. In mobile device, images are greatly affected by environmental light sources. Therefore, obtained image might not qualify for the fingerprint recognition system, even when the image is focused. Consequently, image should be investigated whether it is usable or not. Variance-modified-Laplacian of Gaussian (VMLOG) and image quality index (QI) proposed in this paper solve such problem. VMLOG considers high-frequency component and repeatable patterns of ridges. Experimental results shows that the processing time of the proposed algorithm is enough fast to be adapted in real time system as like mobile device and the proposed algorithm selects exactly a recognizable image.","PeriodicalId":206458,"journal":{"name":"Fourth IEEE Workshop on Automatic Identification Advanced Technologies (AutoID'05)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128666249","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}
The electronic design requirements and applications for a UHF RFID tag emulator are presented. As motivated by present-day industry needs, several implementations of UHF RFID tag emulators are discussed with particular focus given to the use of the emulators as a general-purpose testing tool for RFID system design and on-site measurements. Several noteworthy results of UHF system testing are mentioned. As a longer-term development, the emerging use of UHF RFID protocols for general-purpose wireless data communications is also discussed.
{"title":"Design of UHF RFID emulators with applications to RFID testing and data transport","authors":"Richard Redemske, R. Fletcher","doi":"10.1109/AUTOID.2005.18","DOIUrl":"https://doi.org/10.1109/AUTOID.2005.18","url":null,"abstract":"The electronic design requirements and applications for a UHF RFID tag emulator are presented. As motivated by present-day industry needs, several implementations of UHF RFID tag emulators are discussed with particular focus given to the use of the emulators as a general-purpose testing tool for RFID system design and on-site measurements. Several noteworthy results of UHF system testing are mentioned. As a longer-term development, the emerging use of UHF RFID protocols for general-purpose wireless data communications is also discussed.","PeriodicalId":206458,"journal":{"name":"Fourth IEEE Workshop on Automatic Identification Advanced Technologies (AutoID'05)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114585259","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：