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. 发明申请

US20160189000A1 SCALABLE IMAGE MATCHING 审中-公开
标题翻译：可图像匹配
公开(公告)号：US20160189000A1
公开(公告)日：2016-06-30
申请号：US15063050
申请日：2016-03-07
申请人： A9.com, Inc.
发明人： SIMANT DUBE , SUNIL RAMESH , XIAOFAN LIN , ARNAB SANAT KUMAR DHUA , COLIN JON TAYLOR , JAISHANKER K. PILLAI
IPC分类号： G06K9/62
CPC分类号： G06K9/6215 , G06F17/30247 , G06K9/00523 , G06K9/4676 , G06K9/52 , G06K9/6206 , G06K9/6211 , G06K9/6218 , G06K9/6232 , G06K9/6256 , G06K9/6267 , G06K9/6269 , G06K9/6276 , G06K9/6277 , G06K9/6284 , G06K9/66 , G06K2209/19 , H04N19/426 , H04N19/90
摘要： Various embodiments may increase scalability of image representations stored in a database for use in image matching and retrieval. For example, a system providing image matching can obtain images of a number of inventory items, extract features from each image using a feature extraction algorithm, and transform the same into their feature descriptor representations. These feature descriptor representations can be subsequently stored and used to compare against query images submitted by users. Though the size of each feature descriptor representation isn't particularly large, the total number of these descriptors requires a substantial amount of storage space. Accordingly, feature descriptor representations are compressed to minimize storage and, in one example, machine learning can be used to compensate for information lost as a result of the compression.
摘要翻译：各种实施例可以增加存储在用于图像匹配和检索的数据库中的图像表示的可扩展性。例如，提供图像匹配的系统可以获得多个库存物品的图像，使用特征提取算法从每个图像中提取特征，并将其转换成它们的特征描述符表示。这些特征描述符表示可随后存储并用于与用户提交的查询图像进行比较。虽然每个特征描述符表示的大小不是特别大，但是这些描述符的总数需要大量的存储空间。因此，压缩特征描述符表示以最小化存储，并且在一个示例中，可以使用机器学习来补偿由于压缩而丢失的信息。

22. 发明授权

US09324007B2 Systems and methods for detecting and coordinating changes in lexical items 有权
标题翻译：用于检测和协调词汇项目变更的系统和方法
公开(公告)号：US09324007B2
公开(公告)日：2016-04-26
申请号：US13598662
申请日：2012-08-30
申请人： Jeremy Wright , Allcia Abella , John Grothendieck
发明人： Jeremy Wright , Allcia Abella , John Grothendieck
IPC分类号： G06F9/44 , G06N7/02 , G06N7/06 , G06K9/62 , G06N5/00 , G06N5/02 , G06F17/30
CPC分类号： G06F17/2735 , G06F17/2785 , G06F17/30516 , G06K9/6284 , G06N5/003 , G06N5/022
摘要： Systems and methods for efficiently detecting and coordinating step changes, trends, cycles, and bursts affecting lexical items within data streams are provided. Data streams can be sourced from documents that can optionally be labeled with metadata. Changes can be grouped across lexical and/or metavalue vocabularies to summarize the changes that are synchronous in time. The methods described herein can be applied either retrospectively to a corpus of data or in a streaming mode.
摘要翻译：提供了有效检测和协调影响数据流中词汇项的步骤变化，趋势，周期和突发的系统和方法。数据流可以来自可以选择用元数据标记的文档。变更可以在词汇和/或元空间词汇中分组，以总结时间上同步的变化。这里描述的方法可以追溯地应用于数据语料库或流媒体模式。

23. 发明申请

US20150339265A1 DETECTING ANOMALIES IN A TIME SERIES 有权
标题翻译：检测时间序列中的异常
公开(公告)号：US20150339265A1
公开(公告)日：2015-11-26
申请号：US14282902
申请日：2014-05-20
申请人： Facebook, Inc.
发明人： Romain J. Thibaux
IPC分类号： G06F17/18
CPC分类号： G06F17/18 , G06K9/00536 , G06K9/6284
摘要： Embodiments are disclosed for detecting anomalies in time series using statistical models. In some embodiments, a linear regression model is built for the time series for predicting future values of the time series. Furthermore, the standard deviation of the difference between a prediction and an ideal value of the time series at any point is then estimated. An anomaly is detected when the difference between the prediction and the observed value is greater than a certain threshold based on the estimated standard deviation.
摘要翻译：公开了使用统计模型检测时间序列异常的实施例。在一些实施例中，为用于预测时间序列的未来值的时间序列建立线性回归模型。此外，估计任何时间序列的预测值与理想值之间的差的标准偏差。当预测和观测值之间的差异大于基于估计的标准偏差的某个阈值时，检测到异常。

24. 发明申请

US20150332126A1 IMAGE PROCESSING APPARATUS, METHOD OF PROCESSING IMAGE, AND PROGRAM 有权
标题翻译：图像处理设备，图像处理方法和程序
公开(公告)号：US20150332126A1
公开(公告)日：2015-11-19
申请号：US14710700
申请日：2015-05-13
申请人： Satoshi HIKIDA
发明人： Satoshi HIKIDA
IPC分类号： G06K9/62 , G06T3/40 , G06K9/00 , G06K9/46 , H04N5/232 , H04N5/235
CPC分类号： G06K9/6267 , G06K9/00288 , G06K9/00684 , G06K9/00986 , G06K9/4652 , G06K9/6284 , G06K2009/366 , G06T3/4046 , G06T2200/21 , H04N5/23212 , H04N5/23229 , H04N5/23296 , H04N5/2353
摘要： An image processing apparatus calculating to determine one category, to which an input image data belongs, among categories, which are provided to classify image data, includes a convolution processing unit which performs a convolution process using a first coefficient used for calculating to determine the one category and previously learnt; total combination processing units which are prepared for each category performs a total combination process for a processing result obtained by the convolution processing unit using a second coefficient used for calculating to determine the one category and previously learnt, and calculates to determine the one category; and a normalization unit which performs normalization of a processing result obtained by the total combination processing unit for each category, wherein the convolution processing unit performs learning of the first coefficient in common with the categories, and the total combination processing unit performs learning of the second coefficient for each category.
摘要翻译：一种图像处理装置，其计算用于确定输入图像数据属于哪一类别，其被提供以对图像数据进行分类，所述图像处理装置包括：卷积处理单元，其使用用于计算的第一系数来执行卷积处理，以确定一个类别和以前学习; 为每个类别准备的总组合处理单元对卷积处理单元使用用于计算的第二系数的处理结果进行总体组合处理，以确定一个类别并事先学习，并计算确定一个类别; 以及归一化单元，其对于每个类别执行由所述总组合处理单元获得的处理结果的归一化，其中所述卷积处理单元执行与所述类别共同的第一系数的学习，并且所述总组合处理单元执行所述第二每个类别的系数。

25. 发明授权

US09092700B2 Method, system and apparatus for determining a subject and a distractor in an image 有权
标题翻译：用于确定图像中的对象和牵引器的方法，系统和装置
公开(公告)号：US09092700B2
公开(公告)日：2015-07-28
申请号：US13710153
申请日：2012-12-10
申请人： CANON KABUSHIKI KAISHA
发明人： Jue Wang , Clement Fredembach
IPC分类号： G06K9/40 , G06K9/78 , G06K9/46 , G06K9/62
CPC分类号： G06K9/78 , G06K9/4671 , G06K9/6284
摘要： A method of identifying a subject and a distractor in a target image is disclosed. The method receives a reference image comprising image content corresponding to image content of the target image. A first saliency map, which defines a distribution of visual attraction values identifying salient regions within the target image, and a second saliency map, which defines a distribution of visual attraction values identifying salient regions within the reference image, are determined. The method compares image content in salient regions of the first saliency map and the second saliency map. The subject is identified by a salient region of the target image sharing image content with a salient region of the reference image. The distractor is identified based on at least one remaining salient region of the target image.
摘要翻译：公开了一种识别目标图像中的对象和牵引器的方法。该方法接收包括与目标图像的图像内容相对应的图像内容的参考图像。确定识别目标图像中的突出区域的视觉吸引值的分布的第一显着图和限定识别参考图像内的突出区域的视觉吸引值的分布的第二显着图。该方法比较第一显着性图的显着区域和第二显着性图的图像内容。被摄体由目标图像共享图像内容的显着区域与参考图像的显着区域识别。基于目标图像的至少一个剩余显着区域来识别干扰物。

26. 发明申请

US20150054824A1 OBJECT DETECTION METHOD, OBJECT DETECTION DEVICE, AND IMAGE PICKUP DEVICE 有权
标题翻译：对象检测方法，对象检测装置和图像拾取装置
公开(公告)号：US20150054824A1
公开(公告)日：2015-02-26
申请号：US14461911
申请日：2014-08-18
申请人： CANON KABUSHIKI KAISHA
发明人： Yong Jiang
IPC分类号： G06K9/00 , G06K9/46 , G06K9/62 , G06T17/00
CPC分类号： G06K9/6284 , G06K9/6256
摘要： An object detection device comprises a specific object detector configured to detect a specific object in an image, a scene model creation unit configured to create a scene model characterizing a background of the specific object in the image, and a filtering unit configured to filter object detection results of the specific object detector using the scene model to determine the specific object, wherein the scene model creation unit comprises a collection unit configured to collect regions other than the specific object to be detected from the image as samples, a feature extraction unit configured to extract first negative feature vectors from the samples, a clustering unit configured to cluster first negative feature vectors into a plurality of feature groups, and a classifier creation unit configured to create first classifiers each for respective one of the feature groups and to create the scene model by combining the first classifiers.
摘要翻译：一种物体检测装置，包括被配置为检测图像中的特定物体的特定对象检测器，被配置为创建表征图像中的特定对象的背景的场景模型的场景模型生成部，以及被配置为过滤物体检测的过滤部特征对象检测器的结果，使用场景模型来确定特定对象，其中场景模型创建单元包括：收集单元，被配置为从图像中收集除特定对象物以外的区域作为样本;特征提取单元，被配置为从样本中提取第一负特征向量，被配置为将第一负特征向量聚集成多个特征组的聚类单元，以及分类器创建单元，其被配置为为相应的一个特征组创建每个分类器，并且创建场景模型通过组合第一个分类器。

27. 发明授权

US08949072B2 Novelty detection with missing parameters 有权
标题翻译：缺少参数的新奇检测
公开(公告)号：US08949072B2
公开(公告)日：2015-02-03
申请号：US13122801
申请日：2009-10-09
申请人： Lionel Tarassenko , Alistair Hann
发明人： Lionel Tarassenko , Alistair Hann
IPC分类号： G06F3/048 , G06K9/62 , A61B5/0205 , A61B5/00 , G06F19/00 , G06K9/00 , A61B5/021 , A61B5/0245 , A61B5/08 , A61B5/145
CPC分类号： G06K9/6284 , A61B5/02055 , A61B5/021 , A61B5/0245 , A61B5/0816 , A61B5/145 , A61B5/6843 , A61B2560/0276 , G06F19/00 , G06F19/36 , G06K9/00536 , G16H50/70
摘要： A method of obtaining a consistent evaluation of the state of the system which has been monitored by measurement of multiple parameters of that system. The multiple parameters are used to calculate a single dimensional value based on the distance between the current state and normal states of the system using a Parzen Windows probability function. Consistent single dimensional values regardless of the dimensionality of the original data set can be obtained by finding a relationship between the single dimensional value and the probability of status of the system. Different relationships are obtained for different dimensionalities of data sets. Sensor malfunction can also be detected by testing the probability of the state implied by measuring all of the available parameters against the probability of the state implied by ignoring different individual ones of the parameters. A significant disparity in the two probabilities indicate possible sensor malfunction.
摘要翻译：通过测量该系统的多个参数来监测系统状态的一致评估的方法。多个参数用于使用Parzen Windows概率函数基于系统的当前状态和正常状态之间的距离来计算单维值。可以通过找到单维值与系统状态概率之间的关系来获得与原始数据集的维度无关的一致的单维值。对于数据集的不同维度，获得不同的关系。也可以通过测量所有可用参数的概率来检测传感器故障，这是通过忽略不同个体参数隐含的状态的概率来检测的。两个概率的显着差异表明可能的传感器故障。

28. 发明申请

US20140064551A1 CORRECTING ANOMALIES IN TERRAIN DATA 审中-公开
标题翻译：校正数据中的异常
公开(公告)号：US20140064551A1
公开(公告)日：2014-03-06
申请号：US13429271
申请日：2012-03-23
申请人： Stefan BIEGGER
发明人： Stefan BIEGGER
IPC分类号： G06K9/40
CPC分类号： G06K9/0063 , G06K9/4633 , G06K9/6284
摘要： Systems and methods for detecting anomalies and correcting errors in terrain data are provided. In some aspects, a method includes receiving a first terrain image. The method also includes automatically detecting a set of anomalies in the first terrain image. The method also includes generating a modified terrain image based on the first terrain image. The modified terrain image includes a visual indication of at least a subset of the set of anomalies in the first terrain image. The method also includes providing the modified terrain image for display. The method also includes receiving an input indicating that at least a portion of at least one anomaly in the set of anomalies includes an error. The method also includes generating a corrected terrain image by automatically correcting the error. The method also includes providing the corrected terrain image.
摘要翻译：提供了用于检测地形数据异常和纠正错误的系统和方法。在一些方面，一种方法包括接收第一地形图像。该方法还包括自动检测第一地形图像中的一组异常。该方法还包括基于第一地形图像生成修改的地形图像。修改的地形图像包括第一地形图像中的异常集合的至少一个子集的视觉指示。该方法还包括提供用于显示的修改的地形图像。该方法还包括接收指示异常组中的至少一个异常的至少一部分包括错误的输入。该方法还包括通过自动校正误差来产生校正的地形图像。该方法还包括提供校正的地形图像。

29. 发明申请

US20140046878A1 METHOD AND SYSTEM FOR DETECTING SOUND EVENTS IN A GIVEN ENVIRONMENT 审中-公开
标题翻译：用于检测环境中的声音事件的方法和系统
公开(公告)号：US20140046878A1
公开(公告)日：2014-02-13
申请号：US13962866
申请日：2013-08-08
申请人： THALES
发明人： Sebastien LECOMTE , Francois CAPMAN , Bertrand RAVERA , Regis LENGELLE , Cedric RICHARD
IPC分类号： G06N99/00
CPC分类号： G06N20/00 , G06K9/6223 , G06K9/6269 , G06K9/6284 , G10L25/27 , G10L25/51
摘要： A method and system for detecting abnormal events in a given environment comprises a model construction step comprising: a) a step of unsupervised initialization of Q groups; b) a step of definition of a model of normality consisting of 1-class SVM classifiers; c) a step of optimum distribution of the audio signals in the Q different groups; d) repetition of the steps b and c until a stop criterion C1, is checked and a model M is obtained; and a step of use of the model(s) M obtained from the construction step comprising the analysis of an unknown audio signal ST assigning a score to a 1-class SVM classifier, and a comparison of all the scores fq obtained using decision rules in order to determine the presence or absence of an anomaly in the audio signal analyzed.
摘要翻译：一种用于在给定环境中检测异常事件的方法和系统包括：模型构建步骤，包括：a）Q组的无监督初始化步骤; b）定义由1级SVM分类器组成的正态模型的步骤; c）在Q个不同组中音频信号的最佳分布的步骤; d）重复步骤b和c直到停止标准C1被检查并获得模型M; 以及使用从构造步骤获得的模型M的步骤，包括分配给1级SVM分类器的分数的未知音频信号ST的分析，以及使用决定规则获得的所有得分fq的比较以确定所分析的音频信号中是否存在异常。

30. 发明授权

US08630962B2 Error detection method and its system for early detection of errors in a planar or facilities 有权
标题翻译：错误检测方法及其系统，用于早期检测平面或设施中的错误
公开(公告)号：US08630962B2
公开(公告)日：2014-01-14
申请号：US13057831
申请日：2009-05-29
申请人： Shunji Maeda , Hisae Shibuya
发明人： Shunji Maeda , Hisae Shibuya
IPC分类号： G06F15/18
CPC分类号： G05B23/0254 , G06K9/00536 , G06K9/6252 , G06K9/6272 , G06K9/6284
摘要： Provided are a method which permits complete training data and data with added errors, and enables the early and accurate discovery of errors in facilities such as a plant, and a system thereof. To achieve the objectives, (1) the behavior of temporal data is observed over time, and the trace is divided into clusters; (2) the divided cluster groups are modeled in sub spaces, and the discrepancy values are calculated as errors candidates; (3) the training data are used (compare, reference, etc.) for reference to determine the state transitions caused by the changes over time, the environmental changes, the maintenance (parts replacement), and the operation states; and (4) the modeling is a sub space method such as regression analysis or projection distance method of every N data removing N data items, (N=0, 1, 2, . . . ) (for example, when N=1, one error data item is considered to have been added, this data is removed, then the modeling is performed), or a local sub space method. Linear fitting in regression analysis is equivalent to the lowest order regression analysis.
摘要翻译：提供了一种允许完整的训练数据和具有附加错误的数据的方法，并且能够及早准确地发现诸如工厂及其系统之类的设施中的错误。为了实现目标，（1）随时间观察时间数据的行为，并将踪迹分为簇; （2）划分的群集组在子空间中建模，差异值计算为错误候选; （3）使用训练数据（比较，参考等）作为参考，确定随时间变化，环境变化，维护（部件更换）和操作状态引起的状态转换; （4）建模是N次数据去除N个数据项（N = 0,1,2，...）的回归分析或投影距离法的子空间法（例如，当N = 1时，一个错误数据项被认为已被添加，该数据被删除，然后进行建模）或本地子空间方法。回归分析中的线性拟合等价于最低阶回归分析。

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