Pub Date : 1992-08-30DOI: 10.1109/ICPR.1992.201819
T. Rauber, A. Steiger-Garção
Handwritten characters are forms. Thus they are accessible to form description methods. The authors present a universal shape characterization method applied to the domain of off-line handwritten characters. The test universe is restricted to the ten Arabic numerals but can be extended to any other characters. The authors point out previous work in the area of 2D shape classification and handwriting recognition. The theoretical concept of the authors' shape descriptor is introduced together with a recipe to apply it in the real case of discrete binary images. Results of classification experiments are presented.<>
{"title":"Shape description by UNL Fourier features-an application to handwritten character recognition","authors":"T. Rauber, A. Steiger-Garção","doi":"10.1109/ICPR.1992.201819","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201819","url":null,"abstract":"Handwritten characters are forms. Thus they are accessible to form description methods. The authors present a universal shape characterization method applied to the domain of off-line handwritten characters. The test universe is restricted to the ten Arabic numerals but can be extended to any other characters. The authors point out previous work in the area of 2D shape classification and handwriting recognition. The theoretical concept of the authors' shape descriptor is introduced together with a recipe to apply it in the real case of discrete binary images. Results of classification experiments are presented.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"22 1","pages":"466-469"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90857109","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 : 1992-08-30DOI: 10.1109/ICPR.1992.201873
E. Edmonds, L. Pan, Stella M. O'Brien
Proposes a new approach for automatic feature extraction from spectrograms, which is an essential component of acoustic-phonetic analysis in automatic continuous speech recognition. The method comprised four levels: segmentation, pattern classification, feature recognition and labelling, and a post-processor. There were three types of patterns: fuzzy, formant and silence. The extracted features included voice bar, stripes, cut-off and transitions of the first four formants. Some techniques are presented, such as two special distortion functions used in segmentation, and a peak-iterate function to detect the stripes feature. This software has been implemented as part of a speech knowledge interface, which was an expert system for speech analysis for speaker-independent, continuous speech recognition. It has been tested with a set of data chosen from a spectrogram database; the correct detection rate for most features was over 89%, and in some cases was as high as 98%.<>
{"title":"Automatic feature extraction from spectrograms for acoustic-phonetic analysis","authors":"E. Edmonds, L. Pan, Stella M. O'Brien","doi":"10.1109/ICPR.1992.201873","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201873","url":null,"abstract":"Proposes a new approach for automatic feature extraction from spectrograms, which is an essential component of acoustic-phonetic analysis in automatic continuous speech recognition. The method comprised four levels: segmentation, pattern classification, feature recognition and labelling, and a post-processor. There were three types of patterns: fuzzy, formant and silence. The extracted features included voice bar, stripes, cut-off and transitions of the first four formants. Some techniques are presented, such as two special distortion functions used in segmentation, and a peak-iterate function to detect the stripes feature. This software has been implemented as part of a speech knowledge interface, which was an expert system for speech analysis for speaker-independent, continuous speech recognition. It has been tested with a set of data chosen from a spectrogram database; the correct detection rate for most features was over 89%, and in some cases was as high as 98%.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"2 1","pages":"701-704"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89911322","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 : 1992-08-30DOI: 10.1109/ICPR.1992.201817
H. Costin
Presents a practical method for a supervised object classification by means of a decision-making approach using fuzzy sets. The unknown object membership function, as well as the distance between the input symbol and the chosen prototypes, are computed. The classification is made according to the input pattern which maximizes the membership function. The insensitivity of the classification algorithms to the pattern size, misalignment, the possibility of non-complete symbols recognition, and identification of the information source, are accomplished.<>
{"title":"On object classification by means of fuzzy sets' theory","authors":"H. Costin","doi":"10.1109/ICPR.1992.201817","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201817","url":null,"abstract":"Presents a practical method for a supervised object classification by means of a decision-making approach using fuzzy sets. The unknown object membership function, as well as the distance between the input symbol and the chosen prototypes, are computed. The classification is made according to the input pattern which maximizes the membership function. The insensitivity of the classification algorithms to the pattern size, misalignment, the possibility of non-complete symbols recognition, and identification of the information source, are accomplished.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"21 1","pages":"458-461"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88131987","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 : 1992-08-30DOI: 10.1109/ICPR.1992.201818
Zheng-He Yao, Li-De Wu
A hybrid structural and statistical pattern recognition approach to detect oil/gas reservoirs in sand/shale sediments is presented in the paper. On the basis of the sand fiducial profile derived from log data and seismic data, a tree-based region-detecting method is used to detect sand layers, and a Marr's-operator-based clustering algorithm is used to find oil/gas reservoirs in the detected sand layers. The ability of the approach is demonstrated by a real-data example.<>
{"title":"A pattern recognition approach to detect oil/gas reservoirs in sand/shale sediments","authors":"Zheng-He Yao, Li-De Wu","doi":"10.1109/ICPR.1992.201818","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201818","url":null,"abstract":"A hybrid structural and statistical pattern recognition approach to detect oil/gas reservoirs in sand/shale sediments is presented in the paper. On the basis of the sand fiducial profile derived from log data and seismic data, a tree-based region-detecting method is used to detect sand layers, and a Marr's-operator-based clustering algorithm is used to find oil/gas reservoirs in the detected sand layers. The ability of the approach is demonstrated by a real-data example.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"51 1","pages":"462-465"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86413484","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 : 1992-08-30DOI: 10.1109/ICPR.1992.201865
Katrine Weisteen Bjerde, T. Taxt
The ability of three different methods to separate nonrotated symbols of fonts with different slope angles is examined. The authors compare the properties of the elliptic Fourier descriptors of Kuhl and Giardina with the Fourier descriptors of Zahn and Roskies and the grid method.<>
{"title":"Font segmentation of non-rotated, printed symbols in the vector representation","authors":"Katrine Weisteen Bjerde, T. Taxt","doi":"10.1109/ICPR.1992.201865","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201865","url":null,"abstract":"The ability of three different methods to separate nonrotated symbols of fonts with different slope angles is examined. The authors compare the properties of the elliptic Fourier descriptors of Kuhl and Giardina with the Fourier descriptors of Zahn and Roskies and the grid method.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"72 1","pages":"669-672"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89432500","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 : 1992-08-30DOI: 10.1109/ICPR.1992.202153
Y. Tang, X. Cheng, L. Tao, C. Suen, M. Talaat, R. Inglese
A method called concentration-contour method is presented. It transforms a compound pattern into an integral one where contour analysis can be used. The concentration-contour method consists of four phases: (1) concentration of pattern, (2) extraction of contour, (3) transformation of numerical features, and (4) classification. The diagonal-diagonal regional projection transformation (DDRPT), which converts a compound pattern into an integral object, has been used. A VLSI architecture to implement the concentration-contour approach has been designed. The time complexity of the method is only O(N) compared with O(N/sup 2/) when a uniprocessor is used.<>
{"title":"VLSI architecture for parallel concentration-contour approach","authors":"Y. Tang, X. Cheng, L. Tao, C. Suen, M. Talaat, R. Inglese","doi":"10.1109/ICPR.1992.202153","DOIUrl":"https://doi.org/10.1109/ICPR.1992.202153","url":null,"abstract":"A method called concentration-contour method is presented. It transforms a compound pattern into an integral one where contour analysis can be used. The concentration-contour method consists of four phases: (1) concentration of pattern, (2) extraction of contour, (3) transformation of numerical features, and (4) classification. The diagonal-diagonal regional projection transformation (DDRPT), which converts a compound pattern into an integral object, has been used. A VLSI architecture to implement the concentration-contour approach has been designed. The time complexity of the method is only O(N) compared with O(N/sup 2/) when a uniprocessor is used.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"57 1","pages":"151-154"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84855117","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 : 1992-08-30DOI: 10.1109/ICPR.1992.201863
L. Goldfarb
The concept of the transformation system was introduced earlier by the author as a basic part of a general model for pattern learning. In this paper, for several formal languages (of various types) the equivalent transformation systems are presented. From these examples one can draw the conclusion that the transformation systems give shorter and more informative structural class descriptions than the formal grammars.<>
{"title":"Transformation systems are more economical and informative class descriptions than formal grammars","authors":"L. Goldfarb","doi":"10.1109/ICPR.1992.201863","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201863","url":null,"abstract":"The concept of the transformation system was introduced earlier by the author as a basic part of a general model for pattern learning. In this paper, for several formal languages (of various types) the equivalent transformation systems are presented. From these examples one can draw the conclusion that the transformation systems give shorter and more informative structural class descriptions than the formal grammars.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"148 1","pages":"660-664"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77367184","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 : 1992-08-30DOI: 10.1109/ICPR.1992.201832
G. Dimauro, S. Impedovo, G. Pirlo
A statistical analysis on the classification of ambiguous characters as accomplished by humans recognizers is presented. For this purpose some statistical measurements to evaluate ambiguity are proposed. The experimental results carried out taking into account handwritten numerals, make evident some important aspects of human behaviour in pattern classification.<>
{"title":"Classification of ambiguous patterns","authors":"G. Dimauro, S. Impedovo, G. Pirlo","doi":"10.1109/ICPR.1992.201832","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201832","url":null,"abstract":"A statistical analysis on the classification of ambiguous characters as accomplished by humans recognizers is presented. For this purpose some statistical measurements to evaluate ambiguity are proposed. The experimental results carried out taking into account handwritten numerals, make evident some important aspects of human behaviour in pattern classification.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"120 1","pages":"520-524"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77439785","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 : 1992-08-30DOI: 10.1109/ICPR.1992.201750
A. Kawamura, K. Yura, T. Hayama, Y. Hidai, T. Minamikawa, A. Tanaka, S. Masuda
The authors propose an online handwritten Japanese character recognition method permitting both stroke number and stroke order variations. The method is based on the pattern matching technique. Matching is done by the multiple similarity method using directional feature densities, which are independent of both stroke number and stroke order. This method has achieved a good recognition rate, 91%, for 2965 freely written Japanese kanji characters.<>
{"title":"Online recognition of freely handwritten Japanese characters using directional feature densities","authors":"A. Kawamura, K. Yura, T. Hayama, Y. Hidai, T. Minamikawa, A. Tanaka, S. Masuda","doi":"10.1109/ICPR.1992.201750","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201750","url":null,"abstract":"The authors propose an online handwritten Japanese character recognition method permitting both stroke number and stroke order variations. The method is based on the pattern matching technique. Matching is done by the multiple similarity method using directional feature densities, which are independent of both stroke number and stroke order. This method has achieved a good recognition rate, 91%, for 2965 freely written Japanese kanji characters.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"23 1","pages":"183-186"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87053833","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 : 1992-08-30DOI: 10.1109/ICPR.1992.202135
P. Chakrapani, A. Khokhar, V. Prasanna
Presents a processor-time optimal partitioned implementation of a stereo matching algorithm using zero crossing points as matching primitives. The authors provide O(nm/P) time algorithm on square root P* square root P processor mesh array, where n is the number of zero crossing points in the left image, m is the set of possible candidate points in the right image for a given zero crossing point, and 1>
{"title":"Parallel stereo on fixed size arrays using zero crossings","authors":"P. Chakrapani, A. Khokhar, V. Prasanna","doi":"10.1109/ICPR.1992.202135","DOIUrl":"https://doi.org/10.1109/ICPR.1992.202135","url":null,"abstract":"Presents a processor-time optimal partitioned implementation of a stereo matching algorithm using zero crossing points as matching primitives. The authors provide O(nm/P) time algorithm on square root P* square root P processor mesh array, where n is the number of zero crossing points in the left image, m is the set of possible candidate points in the right image for a given zero crossing point, and 1<or=P<or=n. The sequential algorithm takes O(nm/sup 2/) time while the faster sequential algorithm shown in the paper runs in O(nm) time.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"52 1","pages":"79-82"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85852564","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：