Pub Date : 2003-09-08DOI: 10.1109/ICDAR.2003.1227700
G. Leedham, Sumit Chachra
The objective of this paper is to present a number of features that can be extracted from handwritten digits and used for author verification or identification of a person's handwriting. The features under consideration are mainly computational features some of which cannot be easily evaluated by humans. On the other hand, these features can be extracted by computer algorithms with a high degree of accuracy. The eleven features used are described. All features were appropriately binarized so that binary feature vectors of constant lengths could be formed. These vectors were then used for author discrimination, using the Hamming distance measure. For this task a writer database consisting of 15 writers was created. Each writer was asked to write random strings of 0 to 9 at least 10 times. The results indicate that the combined features work well at discriminating writers and warrant further detailed investigation. Although the set of features was designed for dealing with handwritten digits (as may be written on cheques), it may also be used for isolated alphabetic characters.
{"title":"Writer identification using innovative binarised features of handwritten numerals","authors":"G. Leedham, Sumit Chachra","doi":"10.1109/ICDAR.2003.1227700","DOIUrl":"https://doi.org/10.1109/ICDAR.2003.1227700","url":null,"abstract":"The objective of this paper is to present a number of features that can be extracted from handwritten digits and used for author verification or identification of a person's handwriting. The features under consideration are mainly computational features some of which cannot be easily evaluated by humans. On the other hand, these features can be extracted by computer algorithms with a high degree of accuracy. The eleven features used are described. All features were appropriately binarized so that binary feature vectors of constant lengths could be formed. These vectors were then used for author discrimination, using the Hamming distance measure. For this task a writer database consisting of 15 writers was created. Each writer was asked to write random strings of 0 to 9 at least 10 times. The results indicate that the combined features work well at discriminating writers and warrant further detailed investigation. Although the set of features was designed for dealing with handwritten digits (as may be written on cheques), it may also be used for isolated alphabetic characters.","PeriodicalId":249193,"journal":{"name":"Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings.","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129252354","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 : 2003-09-08DOI: 10.1109/ICDAR.2003.1227674
Xiaofan Lin
Part-of-speech (POS) tagging is the foundation of natural language processing (NLP) systems, and thus has been an active area of research for many years. However, one question remains unanswered: How will a POS tagger behave when the input text is not error-free? This issue can be of great importance when the text comes from imperfect sources like optical character recognition (OCR). This paper analyzes the performance of both individual POS taggers and combination systems on imperfect text. Experimental results show that a POS tagger's accuracy decreases linearly with the character error rate and the slope indicates a tagger's sensitivity to input text errors.
{"title":"Impact of imperfect OCR on part-of-speech tagging","authors":"Xiaofan Lin","doi":"10.1109/ICDAR.2003.1227674","DOIUrl":"https://doi.org/10.1109/ICDAR.2003.1227674","url":null,"abstract":"Part-of-speech (POS) tagging is the foundation of natural language processing (NLP) systems, and thus has been an active area of research for many years. However, one question remains unanswered: How will a POS tagger behave when the input text is not error-free? This issue can be of great importance when the text comes from imperfect sources like optical character recognition (OCR). This paper analyzes the performance of both individual POS taggers and combination systems on imperfect text. Experimental results show that a POS tagger's accuracy decreases linearly with the character error rate and the slope indicates a tagger's sensitivity to input text errors.","PeriodicalId":249193,"journal":{"name":"Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings.","volume":"93 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120841938","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 : 2003-08-03DOI: 10.1109/ICDAR.2003.1227769
Ondrej Velek, M. Nakagawa
This paper proposes a new approach to acceleratingspeed and increasing the recognition rate of an off-linerecognizer employed for on-line handwriting recognitionof Japanese characters. All training patterns are dividedaccording their stroke number to several groups and onesingle recognizer is dedicated for each group of patterns.Since a number of categories for a single recognizer issmaller, the speed and accuracy improves. First, we makethe model of a recognizer and show that our method cantheoretically accelerate its recognition speed to 45% of theoriginal time. Then, we employ the method to a practicallyused off-line recognizer with the result that the recognitionrate is increased from 90.73% to 91.60% and therecognition time is reduced to only 49.73% of the originalone. Another benefit of our new approach is highscalability so that the recognizer can be optimized forspeed and size or for the best accuracy.
{"title":"Enhancing efficiency and speed of an off-line classifier employed for on-line handwriting recognition of a large character set","authors":"Ondrej Velek, M. Nakagawa","doi":"10.1109/ICDAR.2003.1227769","DOIUrl":"https://doi.org/10.1109/ICDAR.2003.1227769","url":null,"abstract":"This paper proposes a new approach to acceleratingspeed and increasing the recognition rate of an off-linerecognizer employed for on-line handwriting recognitionof Japanese characters. All training patterns are dividedaccording their stroke number to several groups and onesingle recognizer is dedicated for each group of patterns.Since a number of categories for a single recognizer issmaller, the speed and accuracy improves. First, we makethe model of a recognizer and show that our method cantheoretically accelerate its recognition speed to 45% of theoriginal time. Then, we employ the method to a practicallyused off-line recognizer with the result that the recognitionrate is increased from 90.73% to 91.60% and therecognition time is reduced to only 49.73% of the originalone. Another benefit of our new approach is highscalability so that the recognizer can be optimized forspeed and size or for the best accuracy.","PeriodicalId":249193,"journal":{"name":"Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings.","volume":"517 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123101364","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 : 2003-08-03DOI: 10.1109/ICDAR.2003.1227771
H. Fujisawa, Cheng-Lin Liu
Directional features have been successfully used forthe recognition of both machine-printed and handwrittenKanji characters for the last decade. This paper attemptsto explain why the directional features are effective. First,the advances of directional features and related methodsare briefly reviewed. Then the properties that thesimilarity measure should hold are discussed andsimulation experiments of directional pattern matchingare conducted to validate the properties. This analysis isexpected to inspire the design of new and more effectivefeatures.
{"title":"Directional pattern matching for character recognition revisited","authors":"H. Fujisawa, Cheng-Lin Liu","doi":"10.1109/ICDAR.2003.1227771","DOIUrl":"https://doi.org/10.1109/ICDAR.2003.1227771","url":null,"abstract":"Directional features have been successfully used forthe recognition of both machine-printed and handwrittenKanji characters for the last decade. This paper attemptsto explain why the directional features are effective. First,the advances of directional features and related methodsare briefly reviewed. Then the properties that thesimilarity measure should hold are discussed andsimulation experiments of directional pattern matchingare conducted to validate the properties. This analysis isexpected to inspire the design of new and more effectivefeatures.","PeriodicalId":249193,"journal":{"name":"Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings.","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127454514","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 : 2003-08-03DOI: 10.1109/ICDAR.2003.1227630
Huaigu Cao, Xiaoqing Ding, Changsong Liu
A model based approach for rectifying the camera image of the bound document has been developed, i.e., the surface of the document is represented by a general cylindrical surface. The principle of using the model to unwrap the image is discussed. Practically, the skeleton of each horizontal text line is extracted to help estimate the parameter of the model, and rectify the images. To use the model, only a few priori is required, and no more auxiliary device is necessary. Experiment results are given to demonstrate the feasibility and the stability of the method.
{"title":"Rectifying the bound document image captured by the camera: a model based approach","authors":"Huaigu Cao, Xiaoqing Ding, Changsong Liu","doi":"10.1109/ICDAR.2003.1227630","DOIUrl":"https://doi.org/10.1109/ICDAR.2003.1227630","url":null,"abstract":"A model based approach for rectifying the camera image of the bound document has been developed, i.e., the surface of the document is represented by a general cylindrical surface. The principle of using the model to unwrap the image is discussed. Practically, the skeleton of each horizontal text line is extracted to help estimate the parameter of the model, and rectify the images. To use the model, only a few priori is required, and no more auxiliary device is necessary. Experiment results are given to demonstrate the feasibility and the stability of the method.","PeriodicalId":249193,"journal":{"name":"Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings.","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123290958","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 : 2003-08-03DOI: 10.1109/ICDAR.2003.1227738
X. Hilaire
This paper deals with performance evaluation of raster-to-vector conversion algorithms. We briefly review the past work, and focus our attention on a method proposed by Phillips and Chhabra (1999). We show that the matching algorithm used in their method can be improved by introducing new metrics and reformulating the problem as a combinatorial optimization scheme. Improvements brought by our approach are the reduction of the number of utilized thresholds to one, and a gain of stability in the obtained results. We also propose an algorithm to solve the scheme, and compare the evaluation results obtained with both methods on real CAD drawings.
{"title":"A matching scheme to enhance performance evaluation of raster-to-vector conversion algorithms","authors":"X. Hilaire","doi":"10.1109/ICDAR.2003.1227738","DOIUrl":"https://doi.org/10.1109/ICDAR.2003.1227738","url":null,"abstract":"This paper deals with performance evaluation of raster-to-vector conversion algorithms. We briefly review the past work, and focus our attention on a method proposed by Phillips and Chhabra (1999). We show that the matching algorithm used in their method can be improved by introducing new metrics and reformulating the problem as a combinatorial optimization scheme. Improvements brought by our approach are the reduction of the number of utilized thresholds to one, and a gain of stability in the obtained results. We also propose an algorithm to solve the scheme, and compare the evaluation results obtained with both methods on real CAD drawings.","PeriodicalId":249193,"journal":{"name":"Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings.","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125535174","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 : 2003-08-03DOI: 10.1109/ICDAR.2003.1227794
T. Shimamura, Bilan Zhu, Atsushi Masuda, M. Onuma, Takeshi Sakurada, M. Nakagawa, Yasushi Kuronuma
This paper describes prototyping of a form processingsystem employing dot texture for printing input frames ofthe form. The dot texture is the texture composed of smalldots. It eases the separation of handwritings from theinput frames even under monochrome printing/readingenvironments and makes the system to process thehandwritings according to the information embedded inthe dot texture of the frames. The embedded informationin the form dictates how to process the form so that wecall the form "active form" being opposite to the passiveform processed by the program stored in a documentreader. This method can also be used to embed otherinformation such as attribute of handwriting and so on.This paper presents the design, prototyping and somepreliminary evaluation.
{"title":"A prototype of an active form system","authors":"T. Shimamura, Bilan Zhu, Atsushi Masuda, M. Onuma, Takeshi Sakurada, M. Nakagawa, Yasushi Kuronuma","doi":"10.1109/ICDAR.2003.1227794","DOIUrl":"https://doi.org/10.1109/ICDAR.2003.1227794","url":null,"abstract":"This paper describes prototyping of a form processingsystem employing dot texture for printing input frames ofthe form. The dot texture is the texture composed of smalldots. It eases the separation of handwritings from theinput frames even under monochrome printing/readingenvironments and makes the system to process thehandwritings according to the information embedded inthe dot texture of the frames. The embedded informationin the form dictates how to process the form so that wecall the form \"active form\" being opposite to the passiveform processed by the program stored in a documentreader. This method can also be used to embed otherinformation such as attribute of handwriting and so on.This paper presents the design, prototyping and somepreliminary evaluation.","PeriodicalId":249193,"journal":{"name":"Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings.","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115075589","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 : 2003-08-03DOI: 10.1109/ICDAR.2003.1227861
Swapnil Khedekar, V. Ramanaprasad, S. Setlur, V. Govindaraju
In this paper we present a top-down, projection-profilebased algorithm to separate text blocks from image blocksin a Devanagari document. We use a distinctive feature ofDevanagari text, called Shirorekha (Header Line) to analyzethe pattern produced by Devanagari text in the horizontalprofile. The horizontal profile corresponding to a textblock possesses certain regularity in frequency, orientationand shows spatial cohesion. The algorithm uses these featuresto identify text blocks in a document image containingboth text and graphics.
{"title":"Text - image separation in Devanagari documents","authors":"Swapnil Khedekar, V. Ramanaprasad, S. Setlur, V. Govindaraju","doi":"10.1109/ICDAR.2003.1227861","DOIUrl":"https://doi.org/10.1109/ICDAR.2003.1227861","url":null,"abstract":"In this paper we present a top-down, projection-profilebased algorithm to separate text blocks from image blocksin a Devanagari document. We use a distinctive feature ofDevanagari text, called Shirorekha (Header Line) to analyzethe pattern produced by Devanagari text in the horizontalprofile. The horizontal profile corresponding to a textblock possesses certain regularity in frequency, orientationand shows spatial cohesion. The algorithm uses these featuresto identify text blocks in a document image containingboth text and graphics.","PeriodicalId":249193,"journal":{"name":"Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings.","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129593208","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 : 2003-08-03DOI: 10.1109/ICDAR.2003.1227663
S. Marinai, E. Marino, G. Soda
This paper describes a system for efficient indexing and retrieval of words in collections of document images. The proposed method is based on two main principles: unsupervised prototype clustering, and string encoding for efficient string matching. During indexing, a self organizing map (SOM) is trained so as to cluster together similar symbols (character-like objects) in a sub-set of the documents to be stored. By using the trained SOM the words in the whole collection can be stored and represented with a fixed-length description that can be easily compared in order to score most similar words in response to a user query. The system can be automatically adapted to different languages and font styles. The most appropriate applications are for the processing of old documents (18th and 19th Centuries) where current OCRs have more difficulties. Experimental results describe three application scenarios having various levels of difficulty for current OCR systems.
{"title":"Indexing and retrieval of words in old documents","authors":"S. Marinai, E. Marino, G. Soda","doi":"10.1109/ICDAR.2003.1227663","DOIUrl":"https://doi.org/10.1109/ICDAR.2003.1227663","url":null,"abstract":"This paper describes a system for efficient indexing and retrieval of words in collections of document images. The proposed method is based on two main principles: unsupervised prototype clustering, and string encoding for efficient string matching. During indexing, a self organizing map (SOM) is trained so as to cluster together similar symbols (character-like objects) in a sub-set of the documents to be stored. By using the trained SOM the words in the whole collection can be stored and represented with a fixed-length description that can be easily compared in order to score most similar words in response to a user query. The system can be automatically adapted to different languages and font styles. The most appropriate applications are for the processing of old documents (18th and 19th Centuries) where current OCRs have more difficulties. Experimental results describe three application scenarios having various levels of difficulty for current OCR systems.","PeriodicalId":249193,"journal":{"name":"Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings.","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129894921","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 : 2003-08-03DOI: 10.1109/ICDAR.2003.1227808
R. Guest, S. Chindaro, M. Fairhurst, J. Potter
A method for automatically assessing theconstructional sequence from a neuropsychologicaldrawing task using Hidden Markov Models is presented.We also present a method of extracting and identifyingthe position of individual pen strokes relating toindividual sides of a shape within a drawing to formtraining and testing sequences. Our results from twoexperiments using data from patients with visuo-spatialneglect show the HMM classifier is able to generalise onincorrectly extracted sequences and obtain a diagnosticclassification which can be used alongside other forms ofconventional assessment.
{"title":"Automatic classification of hand drawn geometric shapes using constructional sequence analysis","authors":"R. Guest, S. Chindaro, M. Fairhurst, J. Potter","doi":"10.1109/ICDAR.2003.1227808","DOIUrl":"https://doi.org/10.1109/ICDAR.2003.1227808","url":null,"abstract":"A method for automatically assessing theconstructional sequence from a neuropsychologicaldrawing task using Hidden Markov Models is presented.We also present a method of extracting and identifyingthe position of individual pen strokes relating toindividual sides of a shape within a drawing to formtraining and testing sequences. Our results from twoexperiments using data from patients with visuo-spatialneglect show the HMM classifier is able to generalise onincorrectly extracted sequences and obtain a diagnosticclassification which can be used alongside other forms ofconventional assessment.","PeriodicalId":249193,"journal":{"name":"Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings.","volume":"138 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130376685","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：