Pub Date : 2002-08-06DOI: 10.1109/IWFHR.2002.1030925
N. Ayat, M. Cheriet, C. Suen
We address the problem of optimizing kernel parameters in support vector machine modeling, especially when the number of parameters is greater than one as in polynomial kernels and KMOD, our newly introduced kernel. The present work is an extended experimental study of the framework proposed by Chapelle et al. (2001) for optimizing SVM kernels using an analytic upper bound of the error. However our optimization scheme minimizes an empirical error estimate using a quasi-Newton optimization method. To assess our method, the approach is further used for adapting KMOD, RBF and polynomial kernels on synthetic data and NIST database. The method shows a much faster convergence with satisfactory results in comparison with the simple gradient descent method.
{"title":"Empirical error based optimization of SVM kernels: application to digit image recognition","authors":"N. Ayat, M. Cheriet, C. Suen","doi":"10.1109/IWFHR.2002.1030925","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030925","url":null,"abstract":"We address the problem of optimizing kernel parameters in support vector machine modeling, especially when the number of parameters is greater than one as in polynomial kernels and KMOD, our newly introduced kernel. The present work is an extended experimental study of the framework proposed by Chapelle et al. (2001) for optimizing SVM kernels using an analytic upper bound of the error. However our optimization scheme minimizes an empirical error estimate using a quasi-Newton optimization method. To assess our method, the approach is further used for adapting KMOD, RBF and polynomial kernels on synthetic data and NIST database. The method shows a much faster convergence with satisfactory results in comparison with the simple gradient descent method.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125802046","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 : 2002-08-06DOI: 10.1109/IWFHR.2002.1030895
Qizhi Xu, Jinho Kim, L. Lam, C. Suen
This paper describes an off-line system which recognizes unconstrained handwritten month words extracted from Canadian bank cheques. A segmentation based grapheme level HMM (hidden Markov model) classifier and two multilayer perceptron classifiers with different architectures and different features have been developed in CENPARMI for the recognition of month words. In this paper, a combination method with an effective conditional topology is presented, and the most widely used combination rules including Vote, Sum and Product, are experimented. A new modified Product rule is also proposed, which has produced the best recognition rate of 85.36% when tested on a real-life standard Canadian bank cheque database.
{"title":"Recognition of handwritten month words on bank cheques","authors":"Qizhi Xu, Jinho Kim, L. Lam, C. Suen","doi":"10.1109/IWFHR.2002.1030895","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030895","url":null,"abstract":"This paper describes an off-line system which recognizes unconstrained handwritten month words extracted from Canadian bank cheques. A segmentation based grapheme level HMM (hidden Markov model) classifier and two multilayer perceptron classifiers with different architectures and different features have been developed in CENPARMI for the recognition of month words. In this paper, a combination method with an effective conditional topology is presented, and the most widely used combination rules including Vote, Sum and Product, are experimented. A new modified Product rule is also proposed, which has produced the best recognition rate of 85.36% when tested on a real-life standard Canadian bank cheque database.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"167 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124674020","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 : 2002-08-06DOI: 10.1109/IWFHR.2002.1030950
L. M. Mestetskii, I. Reyer, T. Sederberg
A new approach to the segmentation of handwritten text is presented that is based on approximating a binary raster image with a set of polygons and building a continuous skeleton of those polygons. Polygons and skeletons are then used in extraction of lines, removing of spots and artifacts, extraction of words from lines and extraction of strokes from words.
{"title":"Continuous approach to segmentation of handwritten text","authors":"L. M. Mestetskii, I. Reyer, T. Sederberg","doi":"10.1109/IWFHR.2002.1030950","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030950","url":null,"abstract":"A new approach to the segmentation of handwritten text is presented that is based on approximating a binary raster image with a set of polygons and building a continuous skeleton of those polygons. Polygons and skeletons are then used in extraction of lines, removing of spots and artifacts, extraction of words from lines and extraction of strokes from words.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130114627","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 : 2002-08-06DOI: 10.1109/IWFHR.2002.1030915
Gregory F. Russell, M. Perrone, Yi-Min Chee, A. Ziq
This paper investigates the use of both typed and handwritten queries to retrieve handwritten documents. The recognition-based approach reported here is novel in that it expands documents in a fashion analogous to query expansion: Individual documents are expanded using N-best lists which embody additional statistical information from a hidden Markov model (HMM) based handwriting recognizer used to transcribe each of the handwritten documents. This additional information enables the retrieval methods to be robust to machine transcription errors, retrieving documents which otherwise would be unretrievable. Cross-writer experiments on a database of 10985 words in 108 documents from 108 writers, and within-writer experiments in a probabilistic framework, on a database of 537724 words in 3342 documents from 43 writers, indicate that significant improvements in retrieval performance can be achieved. The second database is the largest database of on-line handwritten documents known to its.
{"title":"Handwritten document retrieval","authors":"Gregory F. Russell, M. Perrone, Yi-Min Chee, A. Ziq","doi":"10.1109/IWFHR.2002.1030915","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030915","url":null,"abstract":"This paper investigates the use of both typed and handwritten queries to retrieve handwritten documents. The recognition-based approach reported here is novel in that it expands documents in a fashion analogous to query expansion: Individual documents are expanded using N-best lists which embody additional statistical information from a hidden Markov model (HMM) based handwriting recognizer used to transcribe each of the handwritten documents. This additional information enables the retrieval methods to be robust to machine transcription errors, retrieving documents which otherwise would be unretrievable. Cross-writer experiments on a database of 10985 words in 108 documents from 108 writers, and within-writer experiments in a probabilistic framework, on a database of 537724 words in 3342 documents from 43 writers, indicate that significant improvements in retrieval performance can be achieved. The second database is the largest database of on-line handwritten documents known to its.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"26 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132359741","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 : 2002-08-06DOI: 10.1109/IWFHR.2002.1030960
Masahiro Tanaka, Yumi Ishino, Hironori Shimada, T. Inoue
This paper proposes the point-wise matching method which is to be used as the pre-processing in online signature verification using only the positional information. After this pre-processing, various dynamical or local features of the signature can be used in verification. The test signature and the model one are to be matched point-wise by applying time-variant linear transformation. Kalman filter and the smoother are used for estimating the time-variant transformation parameters. Numerical experiment shows quite a good performance for real online signatures.
{"title":"DP matching using Kalman filter as pre-processing in on-line signature verification","authors":"Masahiro Tanaka, Yumi Ishino, Hironori Shimada, T. Inoue","doi":"10.1109/IWFHR.2002.1030960","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030960","url":null,"abstract":"This paper proposes the point-wise matching method which is to be used as the pre-processing in online signature verification using only the positional information. After this pre-processing, various dynamical or local features of the signature can be used in verification. The test signature and the model one are to be matched point-wise by applying time-variant linear transformation. Kalman filter and the smoother are used for estimating the time-variant transformation parameters. Numerical experiment shows quite a good performance for real online signatures.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"196 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121278696","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 : 2002-08-06DOI: 10.1109/IWFHR.2002.1030938
Matthias Zimmermann, H. Bunke
This paper investigates the use of three different schemes to optimize the number of states of linear left-to-right hidden Markov models (HMM). In the first method, we describe the fixed length modeling scheme where each character model is assigned the same number of states. The second method considered is the Bakis length modeling where the number of model states is set to a given fraction of the average number of observations of the corresponding character. In the third modeling scheme the number of model states is set to a specified quantile of the corresponding character length histogram. This method is called quantile length modeling. A comparison of different length modeling schemes was carried out with a handwriting recognition system using off-line images of cursively handwritten English words from the IAM database. For the fixed length modeling, a recognition rate of 61% was achieved. Using the Bakis or quantile length modeling the word recognition rates could be improved to over 69%.
{"title":"Hidden Markov model length optimization for handwriting recognition systems","authors":"Matthias Zimmermann, H. Bunke","doi":"10.1109/IWFHR.2002.1030938","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030938","url":null,"abstract":"This paper investigates the use of three different schemes to optimize the number of states of linear left-to-right hidden Markov models (HMM). In the first method, we describe the fixed length modeling scheme where each character model is assigned the same number of states. The second method considered is the Bakis length modeling where the number of model states is set to a given fraction of the average number of observations of the corresponding character. In the third modeling scheme the number of model states is set to a specified quantile of the corresponding character length histogram. This method is called quantile length modeling. A comparison of different length modeling schemes was carried out with a handwriting recognition system using off-line images of cursively handwritten English words from the IAM database. For the fixed length modeling, a recognition rate of 61% was achieved. Using the Bakis or quantile length modeling the word recognition rates could be improved to over 69%.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"388 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132351977","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 : 2002-08-06DOI: 10.1109/IWFHR.2002.1030928
S. Kanoun, A. Ennaji, Y. Lecourtier, A. Alimi
A method for Arabic and Latin text block differentiation for printed and handwritten scripts is proposed. This method is based on a morphological analysis for each script at the text block level and a geometrical analysis at the line and the connected component level. In this paper, we present a brief survey, of existing methods used for scripts differentiation as well as a general characteristics of Arabic and Latin scripts. Then, We describe our method for the differentiation of these last scripts. We finally show two experimental results on two different data sets. 400 text blocks constitute the first one and 335 text blocks compose the second.
{"title":"Script and nature differentiation for Arabic and Latin text images","authors":"S. Kanoun, A. Ennaji, Y. Lecourtier, A. Alimi","doi":"10.1109/IWFHR.2002.1030928","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030928","url":null,"abstract":"A method for Arabic and Latin text block differentiation for printed and handwritten scripts is proposed. This method is based on a morphological analysis for each script at the text block level and a geometrical analysis at the line and the connected component level. In this paper, we present a brief survey, of existing methods used for scripts differentiation as well as a general characteristics of Arabic and Latin scripts. Then, We describe our method for the differentiation of these last scripts. We finally show two experimental results on two different data sets. 400 text blocks constitute the first one and 335 text blocks compose the second.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"187 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114736584","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 : 2002-08-06DOI: 10.1109/IWFHR.2002.1030882
S. Veeramachaneni, H. Fujisawa, Cheng-Lin Liu, G. Nagy
Classifiers that utilize style context in co-occurring patterns increase recognition accuracy. When patterns occur as long isogenous fields, this gain should increase unless negated by parameter estimation errors that increase with field length. We show that our method achieves higher accuracy with longer input fields because it can be trained accurately We also present some ongoing work on simple heuristics to reduce computational complexity of the scheme.
{"title":"Classifying isogenous fields","authors":"S. Veeramachaneni, H. Fujisawa, Cheng-Lin Liu, G. Nagy","doi":"10.1109/IWFHR.2002.1030882","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030882","url":null,"abstract":"Classifiers that utilize style context in co-occurring patterns increase recognition accuracy. When patterns occur as long isogenous fields, this gain should increase unless negated by parameter estimation errors that increase with field length. We show that our method achieves higher accuracy with longer input fields because it can be trained accurately We also present some ongoing work on simple heuristics to reduce computational complexity of the scheme.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122791743","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 : 2002-08-06DOI: 10.1109/IWFHR.2002.1030929
Y. Al-Ohali, M. Cheriet, C. Suen
HMM has been successfully used to model 1D data, e.g. voice signals. Their use to model 2D patterns was not as successful due to a major difficulty, in describing the 2D data using 1D observation sequences. In this paper, we discuss the importance of this issue and present an improved method to extract 1D observations from the dynamics of off-line handwritten words. The method is based on pen trajectory estimation techniques. The paper also includes description of our HMM classifier which allows dynamic termination states to achieve enhanced discriminative power. Experimental results show the applicability and usefulness of the proposed method. As a result of using the termination probability in HMM modeling, the top 1/sup st/ recognition rate increased by 10%.
{"title":"Dynamic observations and dynamic state termination for off-line handwritten word recognition using HMM","authors":"Y. Al-Ohali, M. Cheriet, C. Suen","doi":"10.1109/IWFHR.2002.1030929","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030929","url":null,"abstract":"HMM has been successfully used to model 1D data, e.g. voice signals. Their use to model 2D patterns was not as successful due to a major difficulty, in describing the 2D data using 1D observation sequences. In this paper, we discuss the importance of this issue and present an improved method to extract 1D observations from the dynamics of off-line handwritten words. The method is based on pen trajectory estimation techniques. The paper also includes description of our HMM classifier which allows dynamic termination states to achieve enhanced discriminative power. Experimental results show the applicability and usefulness of the proposed method. As a result of using the termination probability in HMM modeling, the top 1/sup st/ recognition rate increased by 10%.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125349065","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 : 2002-08-06DOI: 10.1109/IWFHR.2002.1030903
Akihito Kitadai, M. Nakagawa
This paper describes a prototype learning algorithm for structured character pattern representation with common sub-patterns shared among multiple character templates for online recognition of handwritten Japanese characters. Although prototype learning algorithms have been proved useful for an unstructured set of features, they have not been presented for structured or hierarchical pattern representation. In this paper, we present cost-free parallel translation without rotation of sub-patterns that negates their location distributions and normalization that reflects feature distributions in raw patterns to the sub-pattern prototypes, and then show that a prototype learning algorithm can be applied to the structured character pattern representation with significant effect.
{"title":"A learning algorithm for structured character pattern representation used in online recognition of handwritten Japanese characters","authors":"Akihito Kitadai, M. Nakagawa","doi":"10.1109/IWFHR.2002.1030903","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030903","url":null,"abstract":"This paper describes a prototype learning algorithm for structured character pattern representation with common sub-patterns shared among multiple character templates for online recognition of handwritten Japanese characters. Although prototype learning algorithms have been proved useful for an unstructured set of features, they have not been presented for structured or hierarchical pattern representation. In this paper, we present cost-free parallel translation without rotation of sub-patterns that negates their location distributions and normalization that reflects feature distributions in raw patterns to the sub-pattern prototypes, and then show that a prototype learning algorithm can be applied to the structured character pattern representation with significant effect.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130641073","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：