专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

EP2793172B1 Image processing apparatus, image processing method and program 有权
公开(公告)号：EP2793172B1
公开(公告)日：2018-09-05
申请号：EP14163293.5
申请日：2014-04-03
申请人： OMRON CORPORATION
发明人： Kato, Yutaka , Kiuchi, Yutaka
IPC分类号： G06K9/62 , G06T7/33 , G06T7/00
CPC分类号： G06T7/001 , G06K9/6206 , G06K9/6255 , G06K2209/19 , G06T7/344 , G06T2207/20044 , G06T2207/30164
摘要： To provide an image processing apparatus capable of permitting local shape deformation of a measuring target in pattern matching using a shape feature of the measuring target. A model is defined by a plurality of first positions on an edge extracted from a model image and a changing direction of the edge in each of the first positions. An image processing apparatus (100) calculates a changing direction of an edge in a second position of an input image corresponding to the first position on the edge of the model image. The image processing apparatus (100) accepts an instruction associated with a permissible value of the changing direction of the edge. The image processing apparatus (100) calculates a similarity degree of the first position and the second position corresponding to the first position based on the accepted instruction, the changing direction of the edge in the first and second position. The image processing apparatus (100) determines whether a specific area in the input image is similar to the model or not based on the calculated similarity degree in the second positions.

2. 发明公开

EP3336765A1 OBJECT RECOGNITION DEVICE, OBJECT RECOGNITION PROGRAM, AND IMAGE SEARCH SERVICE PROVIDING METHOD 审中-公开
公开(公告)号：EP3336765A1
公开(公告)日：2018-06-20
申请号：EP18154505.4
申请日：2007-05-09
申请人： Nikon Corporation
发明人： NITTA, Keiichi
IPC分类号： G06K9/62 , G06F17/30
CPC分类号： G06K9/6255 , G06F17/30265 , G06K9/00664 , G06K9/6282 , G06K9/6807
摘要： An object recognition apparatus is provided with an image input unit, an object dictionary unit, and a processor. The image input unit captures image data. The object dictionary unit stores conditions to specify a type of an object in accordance with a hierarchical structure of the types of the objects. The processor specifies the type of the object imaged in the image data, while narrowing down the conditions to be collated upon performing collation of the conditions sequentially in accordance with hierarchies.

3. 发明公开

EP3332357A1 VISUAL REPRESENTATION LEARNING FOR BRAIN TUMOR CLASSIFICATION 审中-公开
公开(公告)号：EP3332357A1
公开(公告)日：2018-06-13
申请号：EP16750307.7
申请日：2016-07-22
申请人： Siemens Aktiengesellschaft
发明人： BHATTACHARYA, Subhabrata , CHEN, Terrence , KAMEN, Ali , SUN, Shanhui
IPC分类号： G06K9/46 , G06K9/00
CPC分类号： G06K9/00147 , A61B5/0042 , A61B5/0084 , G06K9/4609 , G06K9/4676 , G06K9/6223 , G06K9/6247 , G06K9/6255 , G06K9/6262 , G06K2209/05
摘要： Independent subspace analysis (ISA) is used to learn (42) filter kernels for CLE images in brain tumor classification. Convolution (46) and stacking are used for unsupervised learning (44, 48) with ISA to derive the filter kernels. A classifier is trained (56) to classify CLE brain images based on features extracted using the filter kernels. The resulting filter kernels and trained classifier are used (60, 64) to assist in diagnosis of occurrence of brain tumors during or as part of neurosurgical resection. The classification may assist a physician in detecting whether CLE examined brain tissue is healthy or not and/or a type of tumor.

4. 发明公开

EP3321854A2 IDENTIFICATION METHOD, IDENTIFICATION APPARATUS, CLASSIFIER CREATING METHOD, AND CLASSIFIER CREATING APPARATUS 审中-公开
标题翻译：识别方法，识别装置，分类器创建方法和分类器创建装置
公开(公告)号：EP3321854A2
公开(公告)日：2018-05-16
申请号：EP17201112.4
申请日：2017-11-10
申请人： Panasonic Intellectual Property Corporation of America
发明人： SHODA, Yukie , TANIGAWA, Toru
IPC分类号： G06K9/00 , G01S13/93 , G06K9/62 , G08G1/16
CPC分类号： G06K9/0063 , G06K9/00805 , G06K9/46 , G06K9/6227 , G06K9/6228 , G06K9/6255 , G06K9/6256 , G06K9/626 , G06K9/628 , G06T9/002 , G08G1/09623 , G08G1/166
摘要： An identification method includes acquiring image data, acquiring a photography condition for the image data, selecting a classifier corresponding to the acquired photography condition from a plurality of classifiers, each of which was created for one of a plurality of different photography conditions, and identifying a target in the acquired image data by using the selected classifier.
摘要翻译：一种识别方法包括：获取图像数据;获取图像数据的摄影条件;从多个分类器中选择与获取的摄影条件对应的分类器，每个分类器针对多个不同的摄影条件中的一个创建;以及识别通过使用所选择的分类器来对所获取的图像数据进行目标。

5. 发明公开

EP3204888A4 SPATIAL PYRAMID POOLING NETWORKS FOR IMAGE PROCESSING 审中-公开
标题翻译： NETZWERKE MIT空间金字塔池ZUR BILDVERARBEITUNG
公开(公告)号：EP3204888A4
公开(公告)日：2017-10-04
申请号：EP14903561
申请日：2014-10-09
申请人： MICROSOFT TECHNOLOGY LICENSING LLC
发明人： HE KAIMING , SUN JIAN , ZHANG XIANGYU , REN SHAOQING
IPC分类号： G06K9/46
CPC分类号： G06K9/6255 , G06K9/4609 , G06K9/4628 , G06K9/66 , G06N3/04

6. 发明公开

EP3204871A4 GENERIC OBJECT DETECTION IN IMAGES 审中-公开
标题翻译：在BILDERN中检测GENERISCHER OBJEKTE
公开(公告)号：EP3204871A4
公开(公告)日：2017-08-16
申请号：EP14903538
申请日：2014-10-09
申请人： MICROSOFT TECHNOLOGY LICENSING LLC
发明人： HE KAIMING , SUN JIAN , ZHANG XIANGYU
IPC分类号： G06F17/30 , G06K9/46 , G06T1/40
CPC分类号： G06K9/66 , G06F17/30247 , G06F17/30271 , G06K9/4628 , G06K9/6255

7. 发明公开

EP3063704A4 METHOD AND SYSTEM FOR DOCUMENT DATA EXTRACTION TEMPLATE MANAGEMENT 无效
标题翻译： VERFAHREN UND系统ZUR TEMPLATE-VERWALTUNGFÜRDOKUMENTDATENEXTRAKTION
公开(公告)号：EP3063704A4
公开(公告)日：2017-08-09
申请号：EP14859108
申请日：2014-02-04
申请人： INTUIT INC
发明人： MADHANI SUNIL , SREEPATHY ANU , SHENOY MITHUN U
IPC分类号： G06K9/62 , G06F17/24
CPC分类号： G06F17/3053 , G06F17/243 , G06F17/248 , G06F17/30386 , G06K9/00442 , G06K9/6255

8. 发明公开

EP3120285A1 SOFT GENERATION OF BIOMETRIC CANDIDATES AND REFERENCES BASED ON EMPIRICAL BIT ERROR PROBABILITY 有权
标题翻译：基于实证误码BIOMETRIC候选人和参考软GENERATION
公开(公告)号：EP3120285A1
公开(公告)日：2017-01-25
申请号：EP15710189.0
申请日：2015-03-17
申请人： Koninklijke Philips N.V.
发明人： LINNARTZ, Johan-Paul Marie Gerard
IPC分类号： G06F21/32 , G06K9/62 , G06K9/68
CPC分类号： H04L9/3231 , G06F11/10 , G06F21/32 , G06K9/00087 , G06K9/00288 , G06K9/00597 , G06K9/00892 , G06K9/6255 , H03M13/152 , H03M13/63 , H04L9/0643
摘要： A biometric verification device (100) arranged to compare a reference hash (480) with a verification bit string (420) obtained from a biometric, the biometric verification device comprising: —a candidate bit string generator (130) arranged to generate candidate bit strings (430) from the verification bit string and error probabilities, —a hash unit (140) arranged to apply a cryptographic hash function to said generated candidate bit strings to obtain candidate hashes, —a comparison unit (160) arranged to verify if a candidate hash generated by the hash unit matches a reference hash.
摘要翻译：布置成比较基准散列（480）与来自生物测定获得的验证比特串（420）一种生物特征验证装置（100），生物特征验证装置，包括：布置为生成候选比特串-a候选比特串生成器（130）从验证位串和错误概率（430），布置-a散列单元（140）向所述产生的候选的位串，以获得配置为验证是否候选散列，-a比较单元（160）以应用密码散列函数的候选散列由散列单元生成匹配的参考散列。

9. 发明公开

EP3111373A1 UNSUPERVISED TRAINING FOR AN ATLAS-BASED REGISTRATION 有权转让
标题翻译： UNÜBERWACHTES培训FÜREINE ATLASBASIERTE REGISTRIERUNG
公开(公告)号：EP3111373A1
公开(公告)日：2017-01-04
申请号：EP15714930.3
申请日：2015-02-17
申请人： Koninklijke Philips N.V.
发明人： FRANZ, Astrid Ruth , SCHADEWALDT, Nicole , SCHULZ, Heinrich , BYSTROV, Daniel , VIK, Torbjoern
IPC分类号： G06K9/62 , G06T7/00
CPC分类号： G06T7/32 , G06K9/6255 , G06K2009/2045 , G06K2209/05 , G06T7/35 , G06T2207/10081 , G06T2207/20076 , G06T2207/20081 , G06T2207/30004
摘要： A method includes obtaining a single training image from a set of training images in a data repository. The method further includes generating an initial tissue class atlas based on the obtained single training image. The initial tissue class atlas includes two or more different tissue class images corresponding to two or more different tissue classes. The method further includes registering the remaining training images of the set of training images to the initial tissue class atlas. The method further includes generating a quality metric for each of the registered images. The method further includes evaluating the quality metric of each of the registered image with a predetermined evaluation criterion. The method further includes identifying a sub-set of images from the set of training images that satisfy the evaluation criterion. The method further includes generating a subsequent tissue class atlas based on the identified sub-set of the set of training images.
摘要翻译：一种方法包括从数据存储库中的一组训练图像获得单个训练图像。该方法还包括基于获得的单个训练图像生成初始组织类图集。初始组织类图集包括对应于两个或多个不同组织类别的两个或更多个不同的组织分类图像。该方法还包括将该组训练图像的剩余训练图像记录到初始组织分类图集。该方法还包括为每个登记图像生成质量度量。该方法还包括以预定的评估标准评估每个登记图像的质量度量。该方法还包括从满足评估标准的训练图像集中识别图像的子集。该方法还包括基于所识别的该组训练图像的子集来生成随后的组织分类图集。

10. 发明公开

EP2956908A4 MODEL-BASED MULTI-HYPOTHESIS TARGET TRACKER 审中-公开
标题翻译： MODELLBASIERTE MULTI-HYPOTHESEN-ZIELVERFOLGUNGSVORRICHTUNG
公开(公告)号：EP2956908A4
公开(公告)日：2016-12-28
申请号：EP14751122
申请日：2014-01-29
申请人： INTEL CORP
发明人： FLEISHMAN SHAHAR , LIVNY YOTAM , KLIMER GIDEON , KUTLIROFF GERSHOM
IPC分类号： G06T7/00 , G06K9/00 , G06K9/62 , G06T1/00 , G06T7/20
CPC分类号： G06F3/017 , G06F3/011 , G06F3/0304 , G06K9/00335 , G06K9/6255 , G06T7/251 , G06T2207/10028 , G06T2207/30196

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