Pub Date : 2002-11-07DOI: 10.1109/IWFHR.2002.1030892
A. Nakamura, H. Kawajiri
An approach to bigram-based linguistic processing for online handwriting text recognition is described. A probability of correctness for each recognition result is derived from a feature set which consists of bigram probabilities and recognition scores. Using the probability of correctness, the number of candidates accepted to the post-processing step and the weight value balancing recognition scores with bigram scores are adaptively controlled. The proposed method is evaluated in experiments using the HANDS-kuchibue online handwritten character database. Results show that the method is effective in reducing candidates, improving accuracy, and saving computational costs.
{"title":"Bigram-based post-processing for online handwriting recognition using correctness evaluation","authors":"A. Nakamura, H. Kawajiri","doi":"10.1109/IWFHR.2002.1030892","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030892","url":null,"abstract":"An approach to bigram-based linguistic processing for online handwriting text recognition is described. A probability of correctness for each recognition result is derived from a feature set which consists of bigram probabilities and recognition scores. Using the probability of correctness, the number of candidates accepted to the post-processing step and the weight value balancing recognition scores with bigram scores are adaptively controlled. The proposed method is evaluated in experiments using the HANDS-kuchibue online handwritten character database. Results show that the method is effective in reducing candidates, improving accuracy, and saving computational costs.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125380744","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.2002.1030880
J. Pitrelli, M. Perrone
We apply confidence-scoring techniques to verify the output of a handwriting recognizer. We evaluate a variety of scoring functions, including likelihood ratios and estimated posterior probabilities of correctness, in a postprocessing mode to generate confidence scores at the character or word level. Using the post-processor in conjunction with an HMM-based on-line handwriting recognizer for large-vocabulary word recognition, receiver-operating-characteristic (ROC) curves reveal that our post-processor is able to reject correctly 90% of recognizer errors while only falsely rejecting 33% of correctly-recognized words. For isolated-digit recognition, we achieve a correct rejection rate of 90% while keeping false rejection down to 13%.
{"title":"Confidence modeling for verification post-processing for handwriting recognition","authors":"J. Pitrelli, M. Perrone","doi":"10.1109/IWFHR.2002.1030880","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030880","url":null,"abstract":"We apply confidence-scoring techniques to verify the output of a handwriting recognizer. We evaluate a variety of scoring functions, including likelihood ratios and estimated posterior probabilities of correctness, in a postprocessing mode to generate confidence scores at the character or word level. Using the post-processor in conjunction with an HMM-based on-line handwriting recognizer for large-vocabulary word recognition, receiver-operating-characteristic (ROC) curves reveal that our post-processor is able to reject correctly 90% of recognizer errors while only falsely rejecting 33% of correctly-recognized words. For isolated-digit recognition, we achieve a correct rejection rate of 90% while keeping false rejection down to 13%.","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":"117165435","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.2002.1030900
Alexandre Lemieux, Christian Gagné, M. Parizeau
This paper presents experiments with genetically engineered feature sets for recognition of online handwritten characters. These representations stem from a nondescript decomposition of the character frame into a set of rectangular regions, possibly overlapping each represented by a vector of 7 fuzzy variables. Efficient new feature sets are automatically discovered using genetic programming techniques. Recognition experiments conducted on isolated digits of the Unipen database yield improvements of more than 3% over a previously, manually designed representation where region positions and sizes were fixed.
{"title":"Genetic engineering of handwriting representations","authors":"Alexandre Lemieux, Christian Gagné, M. Parizeau","doi":"10.1109/IWFHR.2002.1030900","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030900","url":null,"abstract":"This paper presents experiments with genetically engineered feature sets for recognition of online handwritten characters. These representations stem from a nondescript decomposition of the character frame into a set of rectangular regions, possibly overlapping each represented by a vector of 7 fuzzy variables. Efficient new feature sets are automatically discovered using genetic programming techniques. Recognition experiments conducted on isolated digits of the Unipen database yield improvements of more than 3% over a previously, manually designed representation where region positions and sizes were fixed.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"106 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":"123414776","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.2002.1030918
Marc Fuentes, S. Garcia-Salicetti, B. Dorizzi
We propose in this work to perform on-line signature verification by the fusion of two complementary verification modules. The first one considers a signature as a sequence of points and models the genuine signatures of a given signer by a Hidden Markov Model (HMM). Forgeries are used to compute a decision threshold. In the second module, global parameters of a signature are the inputs of a two-classes neural network trained for each signer on both the genuine and "other" signatures (genuine signatures of other signers). Fusion of the scores given by these two experts through a Support Vector Machine (SVM), allows improving the results over those of each module, on Philips' Database.
{"title":"On line signature verification: Fusion of a Hidden Markov Model and a neural network via a support vector machine","authors":"Marc Fuentes, S. Garcia-Salicetti, B. Dorizzi","doi":"10.1109/IWFHR.2002.1030918","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030918","url":null,"abstract":"We propose in this work to perform on-line signature verification by the fusion of two complementary verification modules. The first one considers a signature as a sequence of points and models the genuine signatures of a given signer by a Hidden Markov Model (HMM). Forgeries are used to compute a decision threshold. In the second module, global parameters of a signature are the inputs of a two-classes neural network trained for each signer on both the genuine and \"other\" signatures (genuine signatures of other signers). Fusion of the scores given by these two experts through a Support Vector Machine (SVM), allows improving the results over those of each module, on Philips' Database.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"29 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":"115039355","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.2002.1030878
Iratxe Soraluze Arriola, Clemente Rodríguez Lafuente, F. Boto, A. Pérez
This paper analyses the application of multistage classifiers based on the k-NN rule to the automatic classification of handwritten digits. The discriminating capacity of a k-NN classifier increases as the size and dimensionality of the reference pattern set (RPS) increases. This supposes a problem for k-NN classifiers in real applications: the high computational cost required. In order to accelerate the process of calculating the distance to each pattern of the RPS, some authors propose the use of condensing techniques. These methods try to reduce the size of the RPS without losing classification power. Our alternative proposal is based on hierarchical classifiers with rejection techniques and incremental learning that reduce the computational cost of the classifier. We have used 270,000 digits (160,000 digits for training and 110, 000 for the test) of the NIST Special Data Bases 19 and 3 (SD19 and SD3) as experimental data sets. The best non -hierarchical classifier achieves a hit rate of 99.50%. The hierarchical classifier achieves the same hit ratio, but with 24.5 times lower computational cost than best non-hierarchical classifier found in our experimentation and 6 times lower than Hart's Algorithm.
{"title":"Multidimensional multistage k-NN classifiers for handwritten digit recognition","authors":"Iratxe Soraluze Arriola, Clemente Rodríguez Lafuente, F. Boto, A. Pérez","doi":"10.1109/IWFHR.2002.1030878","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030878","url":null,"abstract":"This paper analyses the application of multistage classifiers based on the k-NN rule to the automatic classification of handwritten digits. The discriminating capacity of a k-NN classifier increases as the size and dimensionality of the reference pattern set (RPS) increases. This supposes a problem for k-NN classifiers in real applications: the high computational cost required. In order to accelerate the process of calculating the distance to each pattern of the RPS, some authors propose the use of condensing techniques. These methods try to reduce the size of the RPS without losing classification power. Our alternative proposal is based on hierarchical classifiers with rejection techniques and incremental learning that reduce the computational cost of the classifier. We have used 270,000 digits (160,000 digits for training and 110, 000 for the test) of the NIST Special Data Bases 19 and 3 (SD19 and SD3) as experimental data sets. The best non -hierarchical classifier achieves a hit rate of 99.50%. The hierarchical classifier achieves the same hit ratio, but with 24.5 times lower computational cost than best non-hierarchical classifier found in our experimentation and 6 times lower than Hart's Algorithm.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"120 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":"122707994","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.2002.1030906
Simon Günter, H. Bunke
The study of multiple classifier systems has become an area of intensive research in pattern recognition. Also in handwriting, recognition, systems combining several classifiers have been investigated. In the paper new methods for the creation of classifier ensembles based on feature selection algorithms are introduced. These new methods are evaluated and compared to existing approaches in the context of handwritten word recognition, using a hidden Markov model recognizer as basic classifier.
{"title":"Creation of classifier ensembles for handwritten word recognition using feature selection algorithms","authors":"Simon Günter, H. Bunke","doi":"10.1109/IWFHR.2002.1030906","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030906","url":null,"abstract":"The study of multiple classifier systems has become an area of intensive research in pattern recognition. Also in handwriting, recognition, systems combining several classifiers have been investigated. In the paper new methods for the creation of classifier ensembles based on feature selection algorithms are introduced. These new methods are evaluated and compared to existing approaches in the context of handwritten word recognition, using a hidden Markov model recognizer as basic classifier.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"150 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":"122765677","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.2002.1030916
Takuma Yamaguchi, Y. Nakano
This paper proposes an experimental character recognition system to recognise place-names written on signboards. Recently, mobile phones with digital cameras and handy digital cameras have be come popular, so we think this system is useful. In experiments, we tested a total of 112 natural scene images with 320 characters. We obtained a correct character recognition rate of 99% and a place-name recognition rate of 98% (with a rejection rate 2%).
{"title":"Extraction of place-name from natural scenes","authors":"Takuma Yamaguchi, Y. Nakano","doi":"10.1109/IWFHR.2002.1030916","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030916","url":null,"abstract":"This paper proposes an experimental character recognition system to recognise place-names written on signboards. Recently, mobile phones with digital cameras and handy digital cameras have be come popular, so we think this system is useful. In experiments, we tested a total of 112 natural scene images with 320 characters. We obtained a correct character recognition rate of 99% and a place-name recognition rate of 98% (with a rejection rate 2%).","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"121 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":"116838601","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.2002.1030937
A. Nosary, T. Paquet, L. Heutte, A. Bensefia
Handwritten text recognition is a problem rarely studied out of specific applications for which lexical knowledge can constrain the vocabulary to a limited one. In the case of handwritten text recognition, additional information can be exploited to characterize the specificity of the writing. This knowledge can help the recognition system to find coherent solutions from both the lexical and the morphological points of view. We present the principles of a handwritten text recognition system based on the online learning of the writer shapes. The proposed scheme is shown to improve the recognition rates on a sample of fifteen writings, unknown to the system.
{"title":"Handwritten text recognition through writer adaptation","authors":"A. Nosary, T. Paquet, L. Heutte, A. Bensefia","doi":"10.1109/IWFHR.2002.1030937","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030937","url":null,"abstract":"Handwritten text recognition is a problem rarely studied out of specific applications for which lexical knowledge can constrain the vocabulary to a limited one. In the case of handwritten text recognition, additional information can be exploited to characterize the specificity of the writing. This knowledge can help the recognition system to find coherent solutions from both the lexical and the morphological points of view. We present the principles of a handwritten text recognition system based on the online learning of the writer shapes. The proposed scheme is shown to improve the recognition rates on a sample of fifteen writings, unknown to the system.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"181 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":"121727035","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.2002.1030905
Ondrej Velek, Stefan Jäger, M. Nakagawa
We propose a technique for normalizing likelihood of multiple classifiers prior to their combination. Our technique takes classifier-specific likelihood characteristics into account and maps them to a common, ideal characteristic allowing fair combination under arbitrary combination schemes. For each classifier, a simple warping process aligns the likelihood with the accumulated recognition rate, so that recognition rate becomes a uniformly increasing function of likelihood. For combining normalized likelihood values, we investigate several elementary combination rules, such as sum-rule or max-rule. We achieved a significant performance gain of more than five percent, compared to the best single recognition rate, showing both the effectiveness of our method for classifier combination and the benefit of combining on-line Japanese character recognition with stroke order and stroke number independent off-line recognition. Moreover, our experiments provide additional empirical evidence for the good performance of the sum rule in comparison with other elementary combination rules, as has already been observed by other research groups.
{"title":"A new warping technique for normalizing likelihood of multiple classifiers and its effectiveness in combined on-line/off-line japanese character recognition","authors":"Ondrej Velek, Stefan Jäger, M. Nakagawa","doi":"10.1109/IWFHR.2002.1030905","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030905","url":null,"abstract":"We propose a technique for normalizing likelihood of multiple classifiers prior to their combination. Our technique takes classifier-specific likelihood characteristics into account and maps them to a common, ideal characteristic allowing fair combination under arbitrary combination schemes. For each classifier, a simple warping process aligns the likelihood with the accumulated recognition rate, so that recognition rate becomes a uniformly increasing function of likelihood. For combining normalized likelihood values, we investigate several elementary combination rules, such as sum-rule or max-rule. We achieved a significant performance gain of more than five percent, compared to the best single recognition rate, showing both the effectiveness of our method for classifier combination and the benefit of combining on-line Japanese character recognition with stroke order and stroke number independent off-line recognition. Moreover, our experiments provide additional empirical evidence for the good performance of the sum rule in comparison with other elementary combination rules, as has already been observed by other research groups.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"212 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":"115802134","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.2002.1030941
S. Kanoun, A. Ennaji, Y. Lecourtier, A. Alimi
An Arabic text analysis system called AABATAS (affixal approach-based Arabic text analysis system) is proposed. AABATAS recognizes and categorizes the words while identifying their morphological and grammatical characteristics. It is based on a new approach for Arabic word recognition called affixal approach. This affixal approach is guided by the structural properties of language. A dynamic decomposition-recognition mechanism is used in our system and leads to generate a set of reliable solutions for each word. This mechanism attempts to identify, the word basic morphemes: the prefix, the infix, the suffix and the root contrary to the existing approaches that are usually based on the recognition of the whole word or the pseudo-word or the letter. In this paper, we briefly present the general characteristics of Arabic texts as well as a succinct survey of the existing approaches used for their recognition. We then describe the structural properties of the Arabic language and the two systems based on these last properties. The first one concerns a word recognition process and the second is devoted to text analysis. We finally show two experimental results; one on a data set of 545 words and another on a text example.
{"title":"Linguistic integration information in the AABATAS Arabic text analysis system","authors":"S. Kanoun, A. Ennaji, Y. Lecourtier, A. Alimi","doi":"10.1109/IWFHR.2002.1030941","DOIUrl":"https://doi.org/10.1109/IWFHR.2002.1030941","url":null,"abstract":"An Arabic text analysis system called AABATAS (affixal approach-based Arabic text analysis system) is proposed. AABATAS recognizes and categorizes the words while identifying their morphological and grammatical characteristics. It is based on a new approach for Arabic word recognition called affixal approach. This affixal approach is guided by the structural properties of language. A dynamic decomposition-recognition mechanism is used in our system and leads to generate a set of reliable solutions for each word. This mechanism attempts to identify, the word basic morphemes: the prefix, the infix, the suffix and the root contrary to the existing approaches that are usually based on the recognition of the whole word or the pseudo-word or the letter. In this paper, we briefly present the general characteristics of Arabic texts as well as a succinct survey of the existing approaches used for their recognition. We then describe the structural properties of the Arabic language and the two systems based on these last properties. The first one concerns a word recognition process and the second is devoted to text analysis. We finally show two experimental results; one on a data set of 545 words and another on a text example.","PeriodicalId":114017,"journal":{"name":"Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition","volume":"77 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":"116357531","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：