专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

21. 发明公开

EP2758938A1 PROCESSES AND SYSTEMS FOR OPTICAL CHARACTER RECOGNITION USING GENERATED FONT TEMPLATES 审中-公开
标题翻译：方法和系统生产的字体卡的光学字符识别BASIS
公开(公告)号：EP2758938A1
公开(公告)日：2014-07-30
申请号：EP12770319.7
申请日：2012-08-30
申请人： General Electric Company
发明人： FERRO, Andrew, Frank
IPC分类号： G06T7/00 , G06K9/62
CPC分类号： G06K9/6255 , G06K2209/01 , G06T7/0004
摘要： Processes and systems for training machine vision systems for use with OCR algorithms to recognize characters. Such a process includes identifying characters to be recognized and individually generating at least a first set of templates for each of the characters. Each template comprises a grid of cells and is generated by selecting certain cells of the grid to define a pattern that correlates to a corresponding one of the characters. Information relating to the templates is then saved on media, from which the information can be subsequently retrieved to regenerate the templates. The templates can be used in an optical character recognition algorithm to recognize at least some of the characters contained in a marking.
摘要翻译：流程和系统的锻炼机器视觉系统的使用与OCR算法来识别字符。这样的方法包括识别字符被认识和至少独立生成第一组模板用于每个字符。每个模板包括小区网格，并且通过选择网格的某些细胞以限定图案做相关于字符中对应的一个生成。相关的信息模板，然后保存在介质中，从该信息随后可以检索到再生的模板。这些模板可以在光学字符识别算法中使用以识别至少一些包含在一个标记的字符。

22. 发明公开

EP2661732A1 VERFAHREN UND ZAHNRESTAURATIONSERMITTLUNGSSYSTEM ZUR ERMITTLUNG VON ZAHNRESTAURATIONEN 有权
标题翻译： VERFAHREN UND ZAHNRESTAURERMERMITTLUNGSSYSTEM ZUR ERMITTLUNG VON ZAHNRESTAURATIONEN
公开(公告)号：EP2661732A1
公开(公告)日：2013-11-13
申请号：EP11794652.5
申请日：2011-12-12
申请人： Straumann Holding AG
发明人： TANK, Martin
IPC分类号： G06T7/00 , A61C13/00 , G06K9/62
CPC分类号： G06F17/5009 , A61C13/0004 , G06K9/6255 , G06T7/30 , G06T2207/30036
摘要： The invention relates to a method for determining virtual tooth restorations on the basis of scanned data (D) of oral structures, in which a model database (DB) is used, which contains a number of parameterized tooth models (M) for each of a variety tooth types, wherein the parameterization is carried out on the basis of model parameters which comprise position and/or shape parameters, and wherein each tooth model (M) is linked with a number of tooth models (M) of the same tooth type (L). For each desired tooth type, an optimal tooth model (M) is determined from the model database (DB) using an iterative method, in which initially at least one start tooth model (M) of the desired tooth type is selected from the model database (DB) and then, starting with said start tooth model (M), during each iteration step (S) a tooth model (M) is tested with respect to a quality value. The tooth model (M) to be currently tested is adapted to the scanned data (D) by varying model parameters for customization purposes, and a quality value is calculated for the customization. In addition, at least one tooth model (M) to be linked to the tooth model (M) to be tested is likewise customized, and a quality value is calculated. On the basis of the calculated quality values, optionally a new tooth model (M) to be tested of the desired tooth type is selected from the model database (DB) for the next iteration step (S). The iteration is terminated when a quality criterion is reached, and finally at least one virtual tooth restoration (R) is determined from the optimal tooth models (M) and scanned data (D). The invention further relates to a method for generating a model database (DB) for use in such a method, to a method for producing or selecting a tooth restoration part, and to a tooth restoration determination system (5) for determining virtual tooth restorations (R).
摘要翻译：本发明描述了一种基于口腔结构的扫描数据（D）确定虚拟牙齿修复体的方法，其中使用包括多个牙齿类型中的每一种的多个参数化牙齿模型的模型数据库（DB），由此参数化是基于包括位置参数和/或形状参数的模型参数进行，并且由此每个牙齿模型（M）与相同牙齿类型的多个牙齿模型（M）连接，并且其中，对于每个期望的牙齿类型，通过迭代方法来确定模型数据库（DB）中的最佳牙齿模型（M），其中从模型数据库（DB）中首先选择期望的齿型的至少一个起始牙齿模型（M），以及随后从该起始齿模型（M）开始，在每个迭代步骤（S）中，关于质量值测试牙齿模型（M），其中对于个体化，将当前被测试的牙齿模型（M）调整为扫描数据（D）由var 为这个个性化计算出ying模型参数和质量值。此外，在测试中与牙齿模型（M）连接的至少一个牙齿模型（M）也是通过模型参数的变化被调整到扫描数据（D）的个体化，并且为该个体化计算另外的质量值。根据计算的质量值，如果需要，则从下一次迭代步骤（S）的模型数据库（DB）中选择所需齿类型的新牙齿模型（M）。迭代在达到质量标准时中断，最后从最佳牙齿模型（M）和扫描数据（D）中确定至少一个虚拟牙齿恢复。本发明还描述了一种生成用于这种方法的模型数据库（DB）的方法，制造或选择牙齿修复部分的方法以及用于确定虚拟牙齿修复体的牙齿修复确定系统（5）。

23. 发明公开

EP2426622A3 Finding low variance regions in document images for generating image anchor templates for content anchoring, data extraction, and document classification 有权
标题翻译：发现低方差文档图像的区域，用于生成图像模板锚用于锚定内容，数据提取和文件分类
公开(公告)号：EP2426622A3
公开(公告)日：2013-05-01
申请号：EP11177531.8
申请日：2011-08-15
申请人： Palo Alto Research Center Incorporated , Xerox Corporation
发明人： Saund, Eric , Sarkar, Prateek , Brito, Alejandro E. , Bern, Marshall W. , Ragnet, Francois
IPC分类号： G06K9/20 , G06K9/62 , G06K9/46
CPC分类号： G06K9/2054 , G06K9/4671 , G06K9/6255 , G06K2209/01
摘要： Methods of generating image anchor templates from low variance regions of document images of a first class are provided. The methods select (102) a document image from the document images of the first class and align (104) the other document images of the first class to the selected document image. Low variance regions are then determined (106) by comparing the aligned document images and the selected document image and used to generate (108) image anchor templates.

24. 发明授权

EP2128793B1 Verfahren und Vorrichtung zum Überprüfen von Druckerzeugnissen, Computerprogramm und Computerprogrammprodukt 有权
标题翻译：用于检查印刷物，计算机程序和计算机程序产品的方法和装置
公开(公告)号：EP2128793B1
公开(公告)日：2012-10-17
申请号：EP08009752.0
申请日：2008-05-28
申请人： Pepperl + Fuchs GmbH
发明人： Rauscher, Benedikt , Caspari, Thomas , Steinmann, Hendrik , Boenick, Rainer
IPC分类号： G06K9/00 , B42C1/12
CPC分类号： G06K9/6255 , B41F33/0036 , B65H43/00 , B65H2511/413 , B65H2511/512 , B65H2515/84 , B65H2553/42 , B65H2801/21 , B65H2220/01 , B65H2220/02

25. 发明公开

EP2461273A2 Method and system for machine-learning based optimization and customization of document similarities calculation 审中-公开
标题翻译：方法和系统基于机器学习和适应文档的相似度计算的优化
公开(公告)号：EP2461273A2
公开(公告)日：2012-06-06
申请号：EP11190076.7
申请日：2011-11-22
申请人： Palo Alto Research Center Incorporated
发明人： Brdiczka, Oliver
IPC分类号： G06K9/20 , G06K9/62
CPC分类号： G06K9/00442 , G06F17/30699 , G06F17/30705 , G06K9/00483 , G06K9/6255
摘要： One embodiment of the present invention provides a system for optimizing and customizing document-similarity calculation. During operation, the system presents (202) a collection of similar documents to a user, collects (204) feedback on the similarity of the documents from the user, generates (206) generic rules for calculating document similarity, and filters (208) documents with customized similarity calculation based on the feedback provided by the user.
摘要翻译：本发明的一个实施例提供用于优化和定制文档相似性计算的系统。在手术过程中，系统呈现（202）类似的文件到用户的集合，收集来自用户的文档的相似性（204）的反馈，基因速率（206）通用计算文档相似性规则和过滤器（208）的文件用定制的相似度计算基于由用户提供的反馈。

26. 发明授权

EP2102794B1 PORTABLER DATENTRÄGER ZUR BIOMETRISCHEN BENUTZERERKENNUNG 有权转让
标题翻译：便携式媒体生物识别用户检测
公开(公告)号：EP2102794B1
公开(公告)日：2012-04-04
申请号：EP07856423.4
申请日：2007-12-06
申请人： Giesecke & Devrient GmbH
发明人： TAGSCHERER, Michael
IPC分类号： G06K9/62
CPC分类号： G06K9/6255 , G06K9/00885 , G06K9/6251
摘要： A portable data storage medium (1) comprises an executable training module (8) which provides, in a memory (4, 5) of the data storage medium (1), a reference model (10; M) for the biometric identification of a user of the data storage medium (1) using an identification module (9), wherein a biometric comparison vector (16) which is within the acceptance range (A) of the reference model (10; M) is accepted by the identification module (9) as originating from the user. In this case, the training module (8) defines the reference model (10; M) using model reference points (P, P0-P4) which are respectively formed from biometric reference vectors (17; N) originating from the user, wherein a new model reference point (P, P0-P4) is inserted into the reference model (10; M) if the associated reference vector (17; N) is outside the acceptance range (A) of the reference model (10; M).

27. 发明公开

EP2401700A1 TRAITEMENT D'UN FLUX DE DONNÉES NUMÉRIQUES 有权转让
标题翻译：数字数据处理电源
公开(公告)号：EP2401700A1
公开(公告)日：2012-01-04
申请号：EP10710374.9
申请日：2010-02-23
申请人： France Telecom
发明人： NATUREL, Xavier , BERRANI, Sid Ahmed
IPC分类号： G06K9/00 , G06K9/62 , G06F17/30 , H04H60/37
CPC分类号： G06K9/00711 , G06F17/30781 , G06K9/6255 , H04H60/375 , H04H2201/90
摘要： The invention relates to a method for processing a digital data stream including sequences and intersequences, wherein said method comprises a step (E1) of cutting the stream into segments on the fly and calculating signatures for said segments, and a step (E2) of detecting intersequences during which the following operations are applied to said stream on the fly: a first detection of intersequences by separation detection; the supply of a reference base (BR) with at least the signatures of the intersequences detected by the first detection; a second detection of intersequences by comparing the signatures of the stream segments obtained during the cutting step (E1) with the signatures of the intersequence segments contained in the reference base (BR).

28. 发明公开

EP1901239A3 Biometrics authentication apparatus 审中-公开
标题翻译：生物识别认证装置
公开(公告)号：EP1901239A3
公开(公告)日：2012-01-04
申请号：EP07016837.2
申请日：2007-08-28
申请人： YAMAHA CORPORATION
发明人： Yoshioka, Yasuo , Kawahara, Takehiko
IPC分类号： G07C9/00
CPC分类号： G07C9/00158 , G06K9/00885 , G06K9/6255
摘要： In a biometrics authentication apparatus, a storage unit stores a dictionary containing biometrics information of a registered person. An information acquisition unit acquires biometrics information of a subject person. An authentication unit authenticates validity of the subject person based on an index representing a degree of similarity between the biometrics information of the dictionary stored in the storage unit and the acquired biometrics information. A dictionary management unit updates the dictionary based on the acquired biometrics information in a first case where the index represents a degree of the similarity higher than a first threshold value, and otherwise creates a new dictionary based on the acquired biometrics information and stores the created new dictionary in the storage unit, in a second case where the index represents a degree of the similarity lower than the first threshold value and higher than a second threshold value which is set lower than the first threshold value.
摘要翻译：在生物识别装置中，存储单元存储包含注册人员的生物测量信息的字典。信息获取单元获取对象人的生物特征信息。认证单元基于表示存储在存储单元中的词典的生物特征信息与所获取的生物特征信息之间的相似度的指标来认证对象人的有效性。字典管理单元在索引表示相似度高于第一阈值的程度的第一情况下基于所获取的生物测量信息来更新字典，否则基于所获取的生物测量信息创建新的字典并将所创建的新的在索引表示相似度低于第一阈值且高于被设置为低于第一阈值的第二阈值的第二情况下，在存储单元中存储字典。

29. 发明公开

EP2372607A2 Scene matching reference data generation system and position measurement system 审中-公开
标题翻译：系统Er ung v on onüen en en ystem ystem ystem ystem ystem ystem ystem ystem
公开(公告)号：EP2372607A2
公开(公告)日：2011-10-05
申请号：EP11153977.1
申请日：2011-02-10
申请人： Aisin AW Co., Ltd.
发明人： Miyajima, Takayuki
IPC分类号： G06K9/00 , G06K9/62
CPC分类号： G06K9/6255 , G01C21/3602 , G06K9/00664 , G06K9/00805
摘要： A scene matching reference data generation system comprises a data input unit (51), a temporary storage unit (52) adapted to temporarily store captured images, an image similarity degree evaluation unit (53B) adapted to evaluate image similarity degrees of the captured images, a processing target data selection unit (54) adapted to select a plurality of the captured images with image similarity degrees equal to or higher than a predetermined degree, a representative image-capturing position determination unit (56) adapted to determine a representative image-capturing position, an image feature point data generation unit (55) adapted to generate image feature point data based on a plurality of processing target captured images, and a reference data generation unit (57) adapted to generate reference data used for scene matching by associating the image feature point data with the representative image-capturing position.
摘要翻译：场景匹配参考数据生成系统包括数据输入单元（51），适于临时存储捕获图像的临时存储单元（52），适于评估所捕获图像的图像相似度的图像相似度评估单元（53B）适于选择具有等于或高于预定程度的图像相似度的多个拍摄图像的处理对象数据选择单元（54），适于确定代表性图像捕获的代表性图像拍摄位置确定单元（56）位置，适于基于多个处理目标捕获图像生成图像特征点数据的图像特征点数据生成单元（55），以及参考数据生成单元（57），用于生成用于场景匹配的参考数据，具有代表性的图像拍摄位置的图像特征点数据。

30. 发明公开

EP2368214A2 SYSTEMS AND METHODS FOR ANALYZING FACIAL EXPRESSIONS, IDENTIFYING INTENT AND TRANSFORMING IMAGES TRHOUGH REVIEW OF FACIAL EXPRESSIONS 审中-公开
标题翻译：用于分析面部表情的系统和方法，意图识别和转换各种图像以评表情
公开(公告)号：EP2368214A2
公开(公告)日：2011-09-28
申请号：EP09799836.3
申请日：2009-12-23
申请人： Matsumoto, David , Ekman, Paul
发明人： Matsumoto, David , Ekman, Paul
IPC分类号： G06K9/00
CPC分类号： G09B7/02 , G06K9/6253 , G06K9/6255
摘要： Methods of analyzing a plurality of facial expressions are disclosed that include: identifying a subject person, utilizing the subject person to create an image of a known target, removing at least one distracter expression from the target image to form a revised target image, and reviewing the revised target image with at least one third party participant to form a final target image. Additional methods of analyzing a plurality of facial expressions include: identifying a subject person, utilizing the subject person to create an image of a known target, digitizing the target image, removing at least one distracter expression from the target image to transform the target image to a revised target image, and reviewing the revised target image with at least one third party participant to transform the revised target image to a final target image. Software for implementing contemplated methods include: a set speed function, a pre-test phase function, an instruction phase function, a practice phase function, and a post-test phase function.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式