专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

31. 发明授权

US08254697B2 Scalable near duplicate image search with geometric constraints 有权
标题翻译：可扩展的几何重复图像搜索与几何约束
公开(公告)号：US08254697B2
公开(公告)日：2012-08-28
申请号：US12364090
申请日：2009-02-02
申请人： Michael Isard , Qifa Ke , Jian Sun , Zhong Wu
发明人： Michael Isard , Qifa Ke , Jian Sun , Zhong Wu
IPC分类号： G06K9/46 , G06K9/54
CPC分类号： G06F17/30247 , G06K9/4676 , G06K9/6202
摘要： Methods are disclosed for finding images from a large corpus of images that at least partially match a query image. The present method makes use of feature detectors to bundle features into local groups or bundles. These bundled features are repeatable and much more discriminative than an individual SIFT feature. Equally importantly, the bundled features provide a flexible representation that allows simple and robust geometric constraints to be efficiently enforced when querying the index.
摘要翻译：公开了用于从至少部分匹配查询图像的大图像的图像中发现图像的方法。本方法利用特征检测器将特征分组到本地组或束中。这些捆绑功能是可重复的，比单独的SIFT功能更具歧视性。同样重要的是，捆绑功能提供了灵活的表示，允许在查询索引时有效执行简单和强大的几何约束。

32. 发明申请

US20120177294A1 IMAGE RETRIEVAL USING DISCRIMINATIVE VISUAL FEATURES 有权
标题翻译：使用分辨率视觉特征的图像检索
公开(公告)号：US20120177294A1
公开(公告)日：2012-07-12
申请号：US12987167
申请日：2011-01-10
申请人： Qifa Ke , Jia Deng , Simon Baker , Michael Isard
发明人： Qifa Ke , Jia Deng , Simon Baker , Michael Isard
IPC分类号： G06K9/48 , G06K9/52 , G06K9/46
CPC分类号： G06F17/30256 , G06F17/30277
摘要： Image search results are obtained by providing weights to visual features to emphasize features corresponding to objects of interest while simultaneously deemphasizing irrelevant or inconsistent features that lead to poor search results. In order to minimize the impact of visual features that are unreliable or irrelevant with respect to the objects of interest in the image, context-dependent weights are provided to detect visual features such that those visual features pertaining to the objects of interest are more heavily weighted than those visual features that pertain to irrelevant or unreliable portions of the image. Visual features may be weighted for images in a searchable database. Training data may be obtained and used in weighting visual features in a query image and, alternatively, in searchable database images.
摘要翻译：通过为视觉特征提供权重来获得图像搜索结果，以强调与感兴趣对象相对应的特征，同时强调导致不良搜索结果的不相关或不一致的特征。为了最小化相对于图像中感兴趣的对象不可靠或不相关的视觉特征的影响，提供上下文相关权重以检测视觉特征，使得与感兴趣对象相关的那些视觉特征被更加重地加权比那些与图像无关或不可靠部分有关的视觉特征。视觉特征可以对可搜索数据库中的图像进行加权。训练数据可以获得并用于加权查询图像中的视觉特征，或者在可搜索的数据库图像中。

33. 发明授权

US07987365B2 Subscription-based computing implemented in hardware of computing device 有权
标题翻译：基于订阅的计算在计算设备的硬件中实现
公开(公告)号：US07987365B2
公开(公告)日：2011-07-26
申请号：US11388291
申请日：2006-03-24
申请人： Andrew David Birrell , Charles P. Thacker , Michael Isard
发明人： Andrew David Birrell , Charles P. Thacker , Michael Isard
IPC分类号： H04L29/00
CPC分类号： H04L9/3234 , G06F21/121 , H04L9/3247
摘要： A subscription-based computing device has hardware and a subscription enforcer implemented in the hardware. The enforcer has an accumulator that accumulates a usage value as the computing device is being used and an expiration value register that stores an expiration value. The enforcer allows the computing device to operate in a subscription mode without hindrance and with full use when the usage value is less than the stored expiration value, and allows the computing device to operate in an expiration mode with hindrance and without full use when the usage value reaches the stored expiration value to signal that the subscription for the computing device has expired.
摘要翻译：基于订阅的计算设备具有在硬件中实现的硬件和订阅执行器。执行者具有累加器，其在使用计算设备时累积使用值，以及存储过期值的到期值寄存器。执行者允许计算设备在使用值小于存储的到期值的情况下以预定的方式无限制地操作并且充分使用，并且允许计算设备在使用中受到阻碍并且不充分使用的期满模式下操作值达到存储的到期值，以表示计算设备的订阅已经过期。

34. 发明申请

US20070078827A1 Searching for information utilizing a probabilistic detector 失效
标题翻译：使用概率检测器搜索信息
公开(公告)号：US20070078827A1
公开(公告)日：2007-04-05
申请号：US11243924
申请日：2005-10-05
申请人： Gaurav Sareen , Mark Manasse , Martin Abadi , Michael Isard
发明人： Gaurav Sareen , Mark Manasse , Martin Abadi , Michael Isard
IPC分类号： G06F17/30
CPC分类号： G06F17/30687
摘要： A probabilistic detector is utilized to query a database. Utilization of a probabilistic detector provides assurance with 100 per cent probability that a search expression in the query is not in the database index. The probabilistic detector is implemented in the form of a Bloom filter. The probabilistic detector is created by hashing expressions in the database index and mapping the resulting hash values into the probabilistic detector. Upon receiving a query, expressions of the query are hashed. The probabilistic detector is queried using these hash values. If the results of querying the probabilistic detector indicate that searched for information may be in the database, the database is not queried. If the results of querying the probabilistic detector indicate that the information may be in the database, the database is queried for the information using the original query. This technique is advantageous in mitigating detrimental effects of denial of service attacks.
摘要翻译：利用概率检测器查询数据库。概率检测器的利用率提供了100％的可能性，即查询中的搜索表达式不在数据库索引中。概率检测器以Bloom滤波器的形式实现。概率检测器由数据库索引中的散列表达式创建，并将生成的散列值映射到概率检测器中。在接收到查询后，查询的表达式将被哈希。使用这些散列值查询概率检测器。如果查询概率检测器的结果表明搜索到的信息可能在数据库中，则不查询数据库。如果查询概率检测器的结果表明信息可能在数据库中，则使用原始查询查询数据库中的信息。这种技术有利于减轻拒绝服务攻击的有害影响。

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