专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

41. 发明申请

US20150139548A1 METHODS AND SYSTEMS FOR RECOGNIZING HANDWRITING IN HANDWRITTEN DOCUMENTS 有权
标题翻译：用于在手写文档中识别手写的方法和系统
公开(公告)号：US20150139548A1
公开(公告)日：2015-05-21
申请号：US14083560
申请日：2013-11-19
申请人： Xerox Corporation
发明人： Shailesh Vaya , Akshayaram Srinivasan , Sai Praneeth Reddy K
IPC分类号： G06K9/00
CPC分类号： G06K9/00456 , G06K9/6264 , G06K9/6292 , G06Q10/063112
摘要： The disclosed embodiments illustrate a method for comparing handwriting in a first electronic document and a second electronic document. The method includes extracting, by one or more processors, one or more segments from a first electronic document and a second electronic document. Each of the one or more segments includes a handwritten text. Thereafter, one or more sets of segments are created from the one or more segments. An information indicating categorization of each segment in a set of segments in one or more categories is received. The information is provided by one or more workers based on the handwriting in each segment. A similarity score based on a count of segments in each of the one or more categories is determined. The similarity score is deterministic of a degree of similarity between the first electronic document and the second electronic document.
摘要翻译：所公开的实施例示出了用于比较第一电子文档和第二电子文档中的笔迹的方法。该方法包括由一个或多个处理器从第一电子文档和第二电子文档提取一个或多个段。一个或多个片段中的每一个包括手写文本。此后，从一个或多个段创建一组或多组段。接收指示在一个或多个类别的一组段中的每个段的分类的信息。该信息由一个或多个基于每个段中的笔迹的工作者提供。确定基于一个或多个类别中的每个中的段的计数的相似性得分。相似性分数是第一电子文档和第二电子文档之间的相似程度的确定性。

42. 发明申请

US20150112571A1 COMPOSITE CONFIDENCE ESTIMATION FOR PREDICTIVE DRIVER ASSISTANT SYSTEMS 有权
标题翻译：用于预测驾驶员辅助系统的复合信心估计
公开(公告)号：US20150112571A1
公开(公告)日：2015-04-23
申请号：US14520531
申请日：2014-10-22
申请人： Honda Research Institute Europe Gmbh
发明人： Jens SCHMÜDDERICH
IPC分类号： B60W50/00 , G06N5/04
CPC分类号： B60W50/0097 , B60W30/08 , B60W30/14 , B60W30/16 , B60W30/18163 , B60W40/04 , B60W2050/0022 , B60W2550/20 , G06K9/00805 , G06K9/6292 , G06N5/04
摘要： The invention relates to a driving assistance system including a prediction subsystem in a vehicle. According to a method aspect of the invention, the method comprises the steps of accepting a set of basic environment representations; allocating a set of basic confidence estimates; associating weights to the basic confidence estimates; calculating a weighted composite confidence estimate for a composite environment representation; and providing the weighted composite confidence estimate as input for an evaluation of a prediction based on the composite environment representation.
摘要翻译：本发明涉及一种包括车辆中的预测子系统的驾驶辅助系统。根据本发明的方法方面，所述方法包括接受一组基本环境表示的步骤; 分配一组基本的置信度估计; 将权重与基本置信度估计相关联; 计算复合环境表示的加权复合置信估计; 以及提供所述加权复合置信度估计作为用于基于所述复合环境表示的预测的评估的输入。

43. 发明授权

US08930288B2 Learning tags for video annotation using latent subtags 有权
标题翻译：使用潜在子标签学习标记视频注释
公开(公告)号：US08930288B2
公开(公告)日：2015-01-06
申请号：US13294483
申请日：2011-11-11
申请人： George D. Toderici , Weilong Yang
发明人： George D. Toderici , Weilong Yang
IPC分类号： G06F15/18 , G06N99/00 , G11B27/10 , G11B27/28 , G06K9/00 , G06F17/30
CPC分类号： G06N99/005 , G06F17/30799 , G06F17/3082 , G06K9/00718 , G06K9/00744 , G06K9/00751 , G06K9/6292 , G11B27/105 , G11B27/28
摘要： A tag learning module trains video classifiers associated with a stored set of tags derived from textual metadata of a plurality of videos, the training based on features extracted from training videos. Each of the tag classifiers is comprised of a plurality of subtag classifiers relating to latent subtags within the tag. The latent subtags can be initialized by clustering cowatch information relating to the videos for a tag. After initialization to identify subtag groups, a subtag classifier can be trained on features extracted from each subtag group. Iterative training of the subtag classifiers can be accomplished by identifying the latent subtags of a training set using the subtag classifiers, then iteratively improving the subtag classifiers by training each subtag classifier with the videos designated as conforming closest to that subtag.
摘要翻译：标签学习模块训练与从多个视频的文本元数据导出的存储的一组标签相关联的视频分类器，该训练基于从训练视频中提取的特征。每个标签分类器由与标签内的潜在子标签相关的多个子标签分类器组成。可以通过聚集与标签的视频相关的批处理信息来初始化潜在子标签。在初始化以识别子标签组之后，可以对从每个子标签组提取的特征进行子标签分类器的训练。子标签分类器的迭代训练可以通过使用子标记分类器识别训练集的潜在子标题来实现，然后通过训练每个子标记分类器，使用指定为最接近该子标签的视频来迭代地改进子标记分类器。

44. 发明授权

US08929629B1 Method and system for image-based ulcer detection 有权
标题翻译：基于图像的溃疡检测方法和系统
公开(公告)号：US08929629B1
公开(公告)日：2015-01-06
申请号：US14262078
申请日：2014-04-25
申请人： Given Imaging Ltd.
发明人： Stas Rozenfeld
IPC分类号： G06K9/00 , G06T7/00 , A61B5/00 , G06K9/62
CPC分类号： G06T7/0012 , A61B5/07 , A61B5/7275 , G06K9/6292 , G06K2209/05 , G06T2207/10068 , G06T2207/20076 , G06T2207/20081 , G06T2207/30092
摘要： A system and method for ulcer detection which may generate a vector of grades including grades indicative of a probability that the image includes an ulcer, for example an ulcer of specific type. For each grade, generating may include finding ulcer candidates within the image, and for each ulcer candidate, building a property vector describing properties of the ulcer candidate and employing a trained classifier to generate the grade from the property vector. The grades may be combined to obtain an indication or score of the probability that the image includes an ulcer.
摘要翻译：一种用于溃疡检测的系统和方法，其可以产生等级的矢量，包括指示图像包括溃疡（例如特定类型的溃疡）的概率的等级。对于每个等级，生成可以包括在图像内发现溃疡候选物，并且对于每个溃疡候选物，构建描述溃疡候选物的性质的性质向量，并且使用经过训练的分类器从属性向量生成等级。可以组合等级以获得图像包括溃疡的概率的指示或得分。

45. 发明申请

US20140241618A1 Combining Region Based Image Classifiers 审中-公开
标题翻译：组合基于区域的图像分类器
公开(公告)号：US20140241618A1
公开(公告)日：2014-08-28
申请号：US13780330
申请日：2013-02-28
申请人： HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P.
发明人： Steven J. Simske , Malgorzata M. Sturgill , Matthew D. Gaubatz , Paul S. Everest , Masoud Zaverehi
IPC分类号： G06K9/62
CPC分类号： G06K9/6292 , G06K9/6262
摘要： Examples disclosed herein relate to combining region based image classifiers. In one implementation, a processor measures correct classification and misclassification levels associated with a first image classifier related to a first image feature region and measures correct classification and misclassification levels associated with a second image classifier related to a second image feature region. The processor may create a combined classifier based on the first image classifier correct classification and misclassification levels and based on the second image classifier correct classification and misclassification levels such that the combined classifier is related to the first image feature region and the second image feature region.
摘要翻译：本文公开的示例涉及组合基于区域的图像分类器。在一个实现中，处理器测量与与第一图像特征区域相关的第一图像分类器相关联的正确分类和错误分类级别，并且测量与与第二图像特征区域相关的第二图像分类器相关联的正确分类和错误分类级别。处理器可以基于第一图像分类器正确分类和错误分类级别并且基于第二图像分类器正确分类和错误分类级别来创建组合分类器，使得组合分类器与第一图像特征区域和第二图像特征区域相关。

46. 发明申请

US20140211014A1 BOUNDARY LINE RECOGNITION APPARATUS 审中-公开
标题翻译：边界线识别装置
公开(公告)号：US20140211014A1
公开(公告)日：2014-07-31
申请号：US14231840
申请日：2014-04-01
申请人： DENSO CORPORATION , NIPPON SOKEN, INC.
发明人： Naoki Kawasaki , Hiroki Nakano , Kenta Hoki , Tetsuya Takafuji
IPC分类号： G06K9/00
CPC分类号： G06K9/00798 , G06K9/6292
摘要： In a boundary line recognition apparatus, a boundary line candidate extracting part extracts boundary line candidates from image data obtained by an on-vehicle camera based on known image processing such as pattern matching and Hough transform. One or more kinds of boundary line feature calculating parts calculate one or more likelihoods of each boundary line candidate. The likelihood indicates a degree of probability to be the boundary line. A boundary line feature combining means multiplies the likelihoods of each boundary line candidate and outputs a combined likelihood. A boundary line candidate selecting part selects the boundary line candidate having a maximum likelihood as the boundary line. The boundary line feature calculating part further calculates the likelihood of the boundary line candidate using a dispersion of brightness and an internal edge amount, and changes the likelihood based on an additional likelihood obtained by a driving lane surface feature extracting part.
摘要翻译：在边界线识别装置中，边界线候选提取部分基于诸如图案匹配和霍夫变换之类的已知图像处理从车载相机获得的图像数据中提取边界线候选。一种或多种边界线特征计算部分计算每个边界线候选的一个或多个似然性。可能性表示作为边界线的概率程度。边界线特征组合装置将每个边界候选候选的可能性相乘并输出组合似然。边界线候补选择部选择具有最大似然度的边界线候选作为边界线。边界线特征计算部分还使用亮度和内部边缘量的色散来计算边界线候选的可能性，并且基于由驾驶车道表面特征提取部获得的附加可能性来改变似然。

47. 发明申请

US20140205187A1 POSE CLASSIFICATION APPARATUS AND METHOD 有权
标题翻译： POSE分类装置和方法
公开(公告)号：US20140205187A1
公开(公告)日：2014-07-24
申请号：US14104524
申请日：2013-12-12
申请人： SAMSUNG ELECTRONICS CO., LTD.
发明人： Ji-hong KANG , Hee-seob RYU , Mikiyas TESHOME , Seung-kwon PARK , Dong-ho LEE , Ki-jun JEONG
IPC分类号： G06K9/62 , G06K9/00
CPC分类号： G06K9/6292 , G06K9/00362
摘要： A pose classification apparatus is provided. The apparatus includes a first image analyzer and a second image analyzer configured to estimate a body part for each pixel of an input image including a human body, a body part decider configured to calculate reliabilities of analysis results of the first image analyzer and the second image analyzer, and configured to decide the body part for each pixel of the input image based on the calculated reliabilities, and a pose estimator configured to estimate a pose of the human body included in the input image, based on the decided body part for each pixel.
摘要翻译：提供姿势分类装置。该装置包括第一图像分析器和第二图像分析器，被配置为估计包括人体的输入图像的每个像素的身体部位，身体部位判定器，被配置为计算第一图像分析器和第二图像的分析结果的可靠性并且被配置为基于所计算的可靠性来决定输入图像的每个像素的身体部分，以及姿势估计器，被配置为基于每个像素的所确定的身体部分来估计输入图像中包括的人体的姿态。

48. 发明授权

US08731295B2 Visual segmentation of lawn grass 有权
标题翻译：草坪草的视觉分割
公开(公告)号：US08731295B2
公开(公告)日：2014-05-20
申请号：US13380413
申请日：2010-07-01
申请人： Alexander Schepelmann , Kathryn A. Daltorio , Amaury D. Rolin , Jonathan Beno , Bradley E. Hughes , James M. Green , Michael S. Branicky , Roger D. Quinn , Henry H. Snow , Francis L. Merat , Richard E. Hudson
发明人： Alexander Schepelmann , Kathryn A. Daltorio , Amaury D. Rolin , Jonathan Beno , Bradley E. Hughes , James M. Green , Michael S. Branicky , Roger D. Quinn , Henry H. Snow , Francis L. Merat , Richard E. Hudson
IPC分类号： G06K9/34
CPC分类号： G06K9/6292 , A01D34/008 , G06K9/4619 , G06K9/4652 , G06T7/11 , G06T7/136 , G06T7/143 , G06T7/44 , G06T7/90 , G06T2207/10024 , G06T2207/20076 , G06T2207/20221 , G06T2207/30188
摘要： Disclosed is a method for identifying lawn grass which includes capturing an image of the terrain in front of a mower, segmenting the image into neighborhoods, calculating at least two image statistics for each of the neighborhoods, generating a binary representation of each image statistic. The binary representation of each image statistic is generated by comparing the calculated image statistic values to predetermined image statistic values for grass. The method further includes weighting each of the binary representations of each image statistic, and summing corresponding neighborhoods for all image statistics. A binary threshold is applied to each of the summed neighborhoods to generate a binary map representing grass containing areas and non-grass containing areas.
摘要翻译：公开了一种用于识别草坪草的方法，其包括在割草机前拍摄地形图像，将图像分割成邻域，计算每个邻域的至少两个图像统计量，生成每个图像统计量的二进制表示。通过将计算的图像统计值与草图的预定图像统计值进行比较，生成每个图像统计量的二进制表示。该方法还包括对每个图像统计量的每个二进制表示进行加权，并且对于所有图像统计信息求和相应的邻域。对每个相加的邻域应用二进制阈值，以生成表示含草区域和非草地区域的二进制图。

49. 发明授权

US08712930B1 Encoding digital content based on models for predicting similarity between exemplars 有权
标题翻译：基于模型来编码数字内容，用于预测样本之间的相似性
公开(公告)号：US08712930B1
公开(公告)日：2014-04-29
申请号：US13100872
申请日：2011-05-04
申请人： Michele Covell , Mei Han , Saurabh Mathur , Shumeet Baluja , Vivek Kwatra
发明人： Michele Covell , Mei Han , Saurabh Mathur , Shumeet Baluja , Vivek Kwatra
IPC分类号： G06F15/18
CPC分类号： G06K9/6202 , G06F17/30247 , G06K9/00684 , G06K9/4604 , G06K9/6267 , G06K9/6292 , G06K9/66 , G06T1/0021 , H04N19/105 , H04N19/17 , H04N19/189
摘要： An exemplar dictionary is built from exemplars of digital content for determining predictor blocks for encoding and decoding digital content. The exemplar dictionary organizes the exemplars as clusters of similar exemplars. Each cluster is mapped to a label. Machine learning techniques are used to generate a prediction model for predicting a label for an exemplar. The exemplar dictionary is used to encode digital content. Clusters of exemplars are obtained by applying a prediction model to a target block of digital content for encoding. A predictor block is selected for encoding the target block based on frequency of occurrence of exemplars in the clusters. The target block is encoded using the predictor block.
摘要翻译：由数字内容的示例构建示范字典，用于确定用于对数字内容进行编码和解码的预测器块。示范字典将样本组织成类似样本的集群。每个集群映射到一个标签。机器学习技术用于生成用于预测样本的标签的预测模型。示范字典用于对数字内容进行编码。通过将预测模型应用于用于编码的数字内容的目标块来获得样本簇。基于群集中的样本的出现频率，选择预测块来对目标块进行编码。使用预测器块对目标块进行编码。

50. 发明申请

US20140110167A1 METHOD AND SYSTEM FOR PREDICTING A DRILL STRING STUCK PIPE EVENT 有权
标题翻译：用于预测钻杆活塞管道事件的方法和系统
公开(公告)号：US20140110167A1
公开(公告)日：2014-04-24
申请号：US13883822
申请日：2012-10-26
申请人： Thomas Goebel , Roberto Valerio Molina , Ricardo Vilalta , Kinjal Dhar Gupta
发明人： Thomas Goebel , Roberto Valerio Molina , Ricardo Vilalta , Kinjal Dhar Gupta
IPC分类号： E21B41/00 , G06N99/00 , G06N3/08 , E21B44/00
CPC分类号： E21B41/0092 , E21B44/00 , G06K9/6292 , G06N3/08 , G06N99/005
摘要： Predicting a drill string stuck pipe event. At least some of the illustrative embodiments are methods including: receiving a plurality of drilling parameters from a drilling operation; applying the plurality of drilling parameters to an ensemble prediction model comprising at least three machine-learning algorithms operated in parallel, each machine-learning algorithm predicting a probability of occurrence of a future stuck pipe event based on at least one of the plurality of drilling parameters, the ensemble prediction model creates a combined probability based on the probability of occurrence of the future stuck pipe event of each machine-learning algorithm; and providing an indication of a likelihood of a future stuck pipe event to a drilling operator, the indication based on the combined probability.
摘要翻译：预测钻柱卡住管道事件。至少一些说明性实施例是一种方法，包括：从钻井操作接收多个钻孔参数; 将所述多个钻孔参数应用于包括并行操作的至少三个机器学习算法的整体预测模型，每个机器学习算法基于所述多个钻孔参数中的至少一个来预测未来卡箍管事件的发生概率综合预测模型基于每个机器学习算法的未来卡通管道事件的发生概率创建组合概率; 并且向钻井操作者提供未来卡住管道事件的可能性的指示，基于组合概率的指示。

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