Pub Date : 1992-08-30DOI: 10.1109/ICPR.1992.202130
K. Araki, M. Shimizu, T. Noda, Y. Chiba, Y. Tsuda, K. Ikegaya, K. Sannomiya, M. Gomi
A new system for high speed and continuous 3-D measurement is presented. It is based on the slit-ray projection method. Remarkable features of the system-high speed and continuous measurement-mainly result from its image plane constructed by a PSD array which is horizontally non-divided and linear, but vertically divided in numbers. All row PSD elements are attached to respective analog signal processors; analog signals are digitized through A/D converters and stored in memory elements in order. By the virtue of this image plane configuration basic data for 3-D measurement are acquired during only one scanning of the slit-ray at high speed in the form of addresses of the memory elements and data stored in them. Moreover, using large capacity and/or dual port memories, one may continuously obtain 3-D range data of as many scenes as one wants. A trial system has been implemented using a newly designed PSD array of 128 channels and it has verified the method, obtaining 3-D range data at video rates continuously in an accuracy within +or-0.3%.<>
{"title":"High speed and continuous 3-D measurement system","authors":"K. Araki, M. Shimizu, T. Noda, Y. Chiba, Y. Tsuda, K. Ikegaya, K. Sannomiya, M. Gomi","doi":"10.1109/ICPR.1992.202130","DOIUrl":"https://doi.org/10.1109/ICPR.1992.202130","url":null,"abstract":"A new system for high speed and continuous 3-D measurement is presented. It is based on the slit-ray projection method. Remarkable features of the system-high speed and continuous measurement-mainly result from its image plane constructed by a PSD array which is horizontally non-divided and linear, but vertically divided in numbers. All row PSD elements are attached to respective analog signal processors; analog signals are digitized through A/D converters and stored in memory elements in order. By the virtue of this image plane configuration basic data for 3-D measurement are acquired during only one scanning of the slit-ray at high speed in the form of addresses of the memory elements and data stored in them. Moreover, using large capacity and/or dual port memories, one may continuously obtain 3-D range data of as many scenes as one wants. A trial system has been implemented using a newly designed PSD array of 128 channels and it has verified the method, obtaining 3-D range data at video rates continuously in an accuracy within +or-0.3%.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"5 1","pages":"62-65"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73367846","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.1992.201786
J. Franke, E. Mandler
Two different approaches are described to combine the results of different classifiers. The first approach is based on the Dempster/Shafer theory of evidence and the second one is a statistical approach with some assumptions on the input data. Both approaches were tested for user-dependent recognition of on-line handwritten characters.<>
{"title":"A comparison of two approaches for combining the votes of cooperating classifiers","authors":"J. Franke, E. Mandler","doi":"10.1109/ICPR.1992.201786","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201786","url":null,"abstract":"Two different approaches are described to combine the results of different classifiers. The first approach is based on the Dempster/Shafer theory of evidence and the second one is a statistical approach with some assumptions on the input data. Both approaches were tested for user-dependent recognition of on-line handwritten characters.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"156 1","pages":"611-614"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73200117","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.1992.201773
S. Satoh, T. Satou, M. Sakauchi
Recently, multi-purpose and generalized image/pattern recognition methods are required. To achieve this, it is desired to realize an automated recognition rule generation system. In this paper, a learning system of structural description rules which are used as models for the model based vision is presented. This system can generate structural description rules efficiently using partially structured example figure descriptions, and a proper inductive inference algorithm for generating structural description rules.<>
{"title":"One method of structural description rule extraction based on graphical and spatial relations","authors":"S. Satoh, T. Satou, M. Sakauchi","doi":"10.1109/ICPR.1992.201773","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201773","url":null,"abstract":"Recently, multi-purpose and generalized image/pattern recognition methods are required. To achieve this, it is desired to realize an automated recognition rule generation system. In this paper, a learning system of structural description rules which are used as models for the model based vision is presented. This system can generate structural description rules efficiently using partially structured example figure descriptions, and a proper inductive inference algorithm for generating structural description rules.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"40 1","pages":"281-284"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74801678","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.1992.201866
Wei Gong, Qinyun Shi, Minde Cheng
New measures for shape and image matching error are presented based on morphological method. It is shown that they grasp the structural and distributive information of the error so as to avoid the incorrect decision by classical ways. The properties of the new measures, such as the dominance of large features, are given. The experiments show these measures have a strong ability against noise and can be used effectively in shape recognition and image registration.<>
{"title":"Shape and image matching by use of morphology","authors":"Wei Gong, Qinyun Shi, Minde Cheng","doi":"10.1109/ICPR.1992.201866","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201866","url":null,"abstract":"New measures for shape and image matching error are presented based on morphological method. It is shown that they grasp the structural and distributive information of the error so as to avoid the incorrect decision by classical ways. The properties of the new measures, such as the dominance of large features, are given. The experiments show these measures have a strong ability against noise and can be used effectively in shape recognition and image registration.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"14 1","pages":"673-676"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74251054","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.1992.201754
C. Higgins, David M. Ford
Users interact with pen-based computers by drawing on the display with a stylus. Handwriting is the obvious method for entering data onto these machines. This requires cursive script recognition (CSR), which comprehends unconstrained, natural handwriting. The authors have developed an online CSR system that can be used by any person with reasonably neat script. This paper describes the segmentation and template matching process used.<>
{"title":"Online recognition of connected handwriting by segmentation and template matching","authors":"C. Higgins, David M. Ford","doi":"10.1109/ICPR.1992.201754","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201754","url":null,"abstract":"Users interact with pen-based computers by drawing on the display with a stylus. Handwriting is the obvious method for entering data onto these machines. This requires cursive script recognition (CSR), which comprehends unconstrained, natural handwriting. The authors have developed an online CSR system that can be used by any person with reasonably neat script. This paper describes the segmentation and template matching process used.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"38 1","pages":"200-203"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74798966","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.1992.202138
G. Tambouratzis, A. Krikelis
The performance of the ASP computer on vision tasks has been evaluated by applying the Abingdon Cross benchmark using a number of different algorithms. In this paper, these algorithms are compared and contrasted on the basis of their performance.<>
{"title":"Implementing the Abingdon Cross benchmark on the ASP","authors":"G. Tambouratzis, A. Krikelis","doi":"10.1109/ICPR.1992.202138","DOIUrl":"https://doi.org/10.1109/ICPR.1992.202138","url":null,"abstract":"The performance of the ASP computer on vision tasks has been evaluated by applying the Abingdon Cross benchmark using a number of different algorithms. In this paper, these algorithms are compared and contrasted on the basis of their performance.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"51 1","pages":"91-94"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75144195","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.1992.201816
C. Lazar
Presents some new results on using the theory of error control codes in pattern recognition. For the shapes described by means of primitive strings , a recognition algorithm is proposed based on string matching. By using the polynomial cyclic code classification, it is obtained the invariance at the position of the starting point when creating the string representation.<>
{"title":"Pattern recognition algorithm based on cyclic codes","authors":"C. Lazar","doi":"10.1109/ICPR.1992.201816","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201816","url":null,"abstract":"Presents some new results on using the theory of error control codes in pattern recognition. For the shapes described by means of primitive strings , a recognition algorithm is proposed based on string matching. By using the polynomial cyclic code classification, it is obtained the invariance at the position of the starting point when creating the string representation.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"66 1","pages":"455-457"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74392359","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.1992.201756
B. S. Manjunath, C. Shekhar, Ramalingam Chellappa, C. Malsburg
The authors present an approach to feature detection, which is a fundamental issue in many intermediate-level vision problems such as stereo, motion correspondence, image registration, etc. The approach is based on a scale-interaction model of the end-inhibition property exhibited by certain cells in the visual- cortex of mammals. These feature detector cells are responsive to short lines, line endings, corners and other such sharp changes in curvature. In addition, this method also provides a compact representation of feature information which is useful in shape recognition problems. Application to face recognition and motion correspondence are illustrated.<>
{"title":"A robust method for detecting image features with application to face recognition and motion correspondence","authors":"B. S. Manjunath, C. Shekhar, Ramalingam Chellappa, C. Malsburg","doi":"10.1109/ICPR.1992.201756","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201756","url":null,"abstract":"The authors present an approach to feature detection, which is a fundamental issue in many intermediate-level vision problems such as stereo, motion correspondence, image registration, etc. The approach is based on a scale-interaction model of the end-inhibition property exhibited by certain cells in the visual- cortex of mammals. These feature detector cells are responsive to short lines, line endings, corners and other such sharp changes in curvature. In addition, this method also provides a compact representation of feature information which is useful in shape recognition problems. Application to face recognition and motion correspondence are illustrated.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"56 1","pages":"208-212"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74625887","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.1992.201867
Xueping Li
Connected or touching numerals appear very often in handwritten documents when they are to be processed by a recognition system. They could be the results of writing habit, limited resolution of the graphic input device, or some preprocessing of the image. In this paper the author describes an approach to identifying connected numerals without a priori knowledge of the number of the numerals in the string. The left boundary of the numeral cage is used for the classification. Special methods are developed for determining the left boundary and extracting the features on it.<>
{"title":"Recognition of connected numeral strings using partial boundary features","authors":"Xueping Li","doi":"10.1109/ICPR.1992.201867","DOIUrl":"https://doi.org/10.1109/ICPR.1992.201867","url":null,"abstract":"Connected or touching numerals appear very often in handwritten documents when they are to be processed by a recognition system. They could be the results of writing habit, limited resolution of the graphic input device, or some preprocessing of the image. In this paper the author describes an approach to identifying connected numerals without a priori knowledge of the number of the numerals in the string. The left boundary of the numeral cage is used for the classification. Special methods are developed for determining the left boundary and extracting the features on it.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"5 1","pages":"677-680"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75027691","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.1992.202145
D. C. Gerogiannis
Issues related to exploitation of parallelism in the perceptual grouping process, a generally accepted paradigm for image segmentation, are explored. The segmentation process is analyzed at three levels: tasks which are performed during a grouping process, computational requirements of algorithms that implement these tasks; and representation of data which are produced/consumed by these algorithms. Based on this analysis, the computational characteristics and requirements of these tasks for parallel implementations are identified and a programming paradigm is suggested for their parallel implementation. It consists of a data representation scheme and a set of primitive operations which have been suggested as basis of languages for data parallel programming. Pseudocode is presented for practical segmentation tasks. The suggested paradigm is compared with other programming models suggested in the literature.<>
{"title":"Programming intermediate level vision tasks on parallel machines","authors":"D. C. Gerogiannis","doi":"10.1109/ICPR.1992.202145","DOIUrl":"https://doi.org/10.1109/ICPR.1992.202145","url":null,"abstract":"Issues related to exploitation of parallelism in the perceptual grouping process, a generally accepted paradigm for image segmentation, are explored. The segmentation process is analyzed at three levels: tasks which are performed during a grouping process, computational requirements of algorithms that implement these tasks; and representation of data which are produced/consumed by these algorithms. Based on this analysis, the computational characteristics and requirements of these tasks for parallel implementations are identified and a programming paradigm is suggested for their parallel implementation. It consists of a data representation scheme and a set of primitive operations which have been suggested as basis of languages for data parallel programming. Pseudocode is presented for practical segmentation tasks. The suggested paradigm is compared with other programming models suggested in the literature.<<ETX>>","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"38 1","pages":"119-123"},"PeriodicalIF":0.0,"publicationDate":"1992-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76346283","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：