专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

51. 发明专利

JP2008171417A Method of detecting substantially rectangular object in image, method of estimating background color in image, computer- readable medium, device for detecting substantially rectangular objects in image, and device for estimating background color in image 审中-公开
标题翻译：检测图像中的大型矩形物体的方法，图像估计方法，计算机可读介质，用于检测图像中的重大矩形物体的装置，以及用于估计图像中的背景颜色的装置
公开(公告)号：JP2008171417A
公开(公告)日：2008-07-24
申请号：JP2007331586
申请日：2007-12-25
申请人： Seiko Epson Corp , セイコーエプソン株式会社
发明人： GUERZHOY MICHAEL , ZHOU HUI
IPC分类号： G06T7/60
CPC分类号： G06T7/143 , G06T7/12 , G06T7/181 , G06T2207/10008 , G06T2207/30176
摘要： PROBLEM TO BE SOLVED: To provide a method and device for detecting objects in an image. SOLUTION: A method of detecting generally rectangular objects in an image includes a step of determining candidate rectangles in the image based on detected corners in the image, a step of ranking the candidate rectangles based on a set of differentiating factors, and a step of detecting objects in the images based on the ranked candidate rectangles. COPYRIGHT: (C)2008,JPO&INPIT
摘要翻译：要解决的问题：提供一种用于检测图像中的物体的方法和装置。解决方案：一种检测图像中的大致矩形物体的方法包括基于图像中检测到的拐角来确定图像中的候选矩形的步骤，基于一组微分因子对候选矩形进行排序的步骤，以及基于排列的候选矩形来检测图像中的对象的步骤。版权所有（C）2008，JPO＆INPIT

52. 发明专利

JP2008058090A Pattern evaluation method and program 有权
标题翻译：模式评估方法与程序
公开(公告)号：JP2008058090A
公开(公告)日：2008-03-13
申请号：JP2006234052
申请日：2006-08-30
申请人： Toshiba Corp , 株式会社東芝
发明人： MITSUI TADASHI
IPC分类号： G01N21/956 , G06T1/00 , H01L21/66
CPC分类号： G06T7/001 , G03F1/84 , G03F1/86 , G06T7/13 , G06T7/181 , G06T2207/10056 , G06T2207/30148
摘要： PROBLEM TO BE SOLVED: To obtain a measuring result of precision equal to that obtained when an image of high magnification is measured, even with respect to an image of low magnification. SOLUTION: The mutual alignment of pattern contours detected from the different inspection images obtained with respect to the same evaluation target pattern is executed and the aligned pattern contours are mutually superposed one upon another to form a single synthetic contour. Then, measurement is made with respect to the obtained synthetic contour. COPYRIGHT: (C)2008,JPO&INPIT
摘要翻译：要解决的问题：为了获得与测量高倍率的图像时相等的精度的测量结果，即使对于低倍率的图像也是如此。解决方案：执行从相同评估目标图案获得的不同检查图像中检测到的图案轮廓的相互对准，并且对齐的图案轮廓彼此相互重叠以形成单个合成轮廓。然后，相对于获得的合成轮廓进行测量。版权所有（C）2008，JPO＆INPIT

53. 发明专利

JP2007325831A Fundus observation apparatus, ophthalmologic image processing apparatus, and ophthalmologic image processing program 有权
标题翻译：基金观察设备，眼科图像处理设备和眼科图像处理程序
公开(公告)号：JP2007325831A
公开(公告)日：2007-12-20
申请号：JP2006160896
申请日：2006-06-09
申请人： Topcon Corp , 株式会社トプコン
发明人： AOKI HIROYUKI , FUJIMURA TAKASHI , FUKUMA YASUFUMI , TSUKADA HIROSHI , KIKAWA TSUTOMU
IPC分类号： A61B3/14 , A61B3/12 , G01N21/17 , G01N21/64
CPC分类号： A61B3/102 , G06T7/12 , G06T7/174 , G06T7/181 , G06T2207/10101 , G06T2207/30041
摘要： PROBLEM TO BE SOLVED: To keep track of a position of an image region corresponding to a cross section of a blood vessel in a tomographic image of a fundus. SOLUTION: A fundus observation apparatus 1 acquires both of a fundus image Ef' of a fundus Ef and tomographic images G1-Gm. An integrated image generation part 231 integrates each tomographic image Gi over the depth direction and generates an integrated image. A blood vessel region selection part 232 selects the blood vessel region respectively from the fundus image Ef' and the integrated image. A blood vessel region positioning part 233 positions the two blood vessel regions. A blood vessel cross sectional region identification part 234 identifies the position of the blood vessel cross sectional region in the each tomographic image Gi from the positioning result. An image interpolation part 235 interpolates images in a layer region in an image region just under the blood vessel cross sectional region (just under the blood vessel region) and a border region. A layer thickness calculation part 236 calculates a thickness of the layer region on a given position of the eyeground Ef including the position just under the blood vessel region. COPYRIGHT: (C)2008,JPO&INPIT
摘要翻译：要解决的问题：跟踪在眼底断层图像中对应于血管横截面的图像区域的位置。解决方案：眼底观察装置1获取眼底Ef的眼底图像Ef'和断层图像G1-Gm两者。集成图像生成部231对深度方向上的各断层图像Gi进行积分，生成集成图像。血管区域选择部件232分别从眼底图像Ef'和集成图像中选择血管区域。血管区域定位部233定位两个血管区域。血管横截面区域识别部件234根据定位结果识别每个断层图像Gi中的血管横截面区域的位置。图像内插部件235在正好在血管横截面区域（血管区域正下方）的图像区域和边界区域中的图像区域内插图像。层厚计算部236计算包括刚好在血管区域下方的位置的眼球Ef的给定位置上的层区域的厚度。版权所有（C）2008，JPO＆INPIT

54. 发明专利

JP2007133874A Method and system for generating polygon boundary definition for image object 有权
标题翻译：用于生成图像对象的多边形定义的方法和系统
公开(公告)号：JP2007133874A
公开(公告)日：2007-05-31
申请号：JP2006301296
申请日：2006-11-07
申请人： General Electric Co , ゼネラル・エレクトリック・カンパニイGeneral Electric Company
发明人： BRUNO JEANETTE MARIE
IPC分类号： G06T3/00 , G06T7/60
CPC分类号： G06K9/48 , G06K9/4638 , G06T7/12 , G06T7/181
摘要： PROBLEM TO BE SOLVED: To provide an improved method for identifying an image object boundary. SOLUTION: When boundary points 26, 27, 50 and 52, which describe lines defining a polygon deciding an outline of each object, are found, an image object 28 inside a two-dimensional image is identified. In processing, lines 32, 34, 36 and 38 of pixels between the boundary points are tracked for determining whether an added pixel 30 is a part of the image object or not by referring to a discriminator. Consequently, the boundary points are identified for contact pixels and contact lines. Arm-shaped areas 74, 76, 78, 79, and 80 are identified by recursive analysis based on the similar method. A list of the consequent boundary points can be used for identifying the image object for labeling/tagging in a web-based manual/catalog. COPYRIGHT: (C)2007,JPO&INPIT
摘要翻译：要解决的问题：提供一种用于识别图像对象边界的改进方法。解决方案：当找到描述确定每个对象的轮廓的多边形的线的边界点26,27,50和52时，识别二维图像内部的图像对象28。在处理中，跟踪边界点之间的像素的线32,34,36和38，以通过参考鉴别器来确定添加的像素30是否是图像对象的一部分。因此，为接触像素和接触线识别边界点。基于类似的方法，通过递归分析来识别手臂区域74,76,78,79和80。随后的边界点的列表可用于识别基于Web的手册/目录中的标签/标记的图像对象。版权所有（C）2007，JPO＆INPIT

55. 发明专利

JP2006297104A Method and system for detecting vessel boundary 审中-公开
标题翻译：检测船舶边界的方法和系统
公开(公告)号：JP2006297104A
公开(公告)日：2006-11-02
申请号：JP2006115697
申请日：2006-04-19
申请人： Siemens Corporate Res Inc , シーメンスコーポレイトリサーチインコーポレイテツドSiemens Corporate Research,Inc.
发明人： TEK HUSEYIN , AYVACI ALPER , DORIN COMANICIU
IPC分类号： A61B5/00 , A61B5/055 , A61B6/03 , G01R33/32 , G06T1/00
CPC分类号： G06T7/60 , G06T7/12 , G06T7/181 , G06T2207/10072 , G06T2207/20116 , G06T2207/20156 , G06T2207/20192 , G06T2207/30101
摘要： PROBLEM TO BE SOLVED: To detect a boundary of vessel correctly and certainly. SOLUTION: The method and system for detecting vessel boundary and a medium capable of reading out from a computer are characterized by the fact that a plurality of edges in the image are detected based on the change in intensity between data points over some distance, and then a set of edges is selected from the plurality of edges, and an initial vessel boundary is determined based on the selected set of edges, and a shape descriptor is applied to the initial vessel boundary to determine a final vessel boundary. COPYRIGHT: (C)2007,JPO&INPIT
摘要翻译：要解决的问题：正确和肯定地检测船只的边界。解决方案：用于检测血管边界的方法和系统以及能够从计算机读出的介质的特征在于，基于在一定距离上的数据点之间的强度变化来检测图像中的多个边缘，然后从多个边缘中选择一组边缘，并且基于所选择的边缘集合来确定初始容器边界，并且将形状描述符应用于初始容器边界以确定最终的容器边界。版权所有（C）2007，JPO＆INPIT

56. 发明专利

KR102226845B1 KR102226845B1 - System and method for object recognition using local binarization 有权
公开(公告)号：KR102226845B1
公开(公告)日：2021-03-12
申请号：KR1020190022779A
申请日：2019-02-26
申请人： 주식회사 핀그램 , 주식회사 큐램
发明人： 양성진 , 위영철 , 안영훈
IPC分类号： G06K9/34 , G06K9/32 , G06K9/38 , G06K9/46
CPC分类号： G06K9/4609 , G06K9/344 , G06K9/3233 , G06K9/325 , G06K9/38 , G06K9/4604 , G06T7/11 , G06T7/12 , G06T7/181 , G06K2009/4666 , G06K2209/01 , G06K2209/21 , G06T2207/10024 , G06T2207/20212
摘要： 지역적 이진화를 이용한 오브젝트 인식방법 및 그 시스템이 개시된다. 상기 오브젝트 인식방법은 오브젝트 인식시스템이 인식대상 오브젝트가 표시된 이미지에서 상기 이미지의 이진화 대상영역의 일부에 제1이진화 영역 및 상기 제1이진화 영역을 포함하며 상기 제1이진화 영역보다 일정 비율 넓은 영역을 가지는 제1계수 연산영역을 설정하는 단계, 상기 오브젝트 인식시스템이 상기 제1계수 연산영역에 포함된 픽셀들의 픽셀 값들에 기초하여 제1이진화 계수를 결정하는 단계, 상기 오브젝트 인식시스템이 결정된 상기 제1이진화 계수를 통해 상기 제1이진화 영역에 대한 이진화를 수행하는 단계, 상기 오브젝트 인식시스템이 상기 제1이진화 영역이 아닌 타영역에 제2이진화 영역 및 상기 제2이진화 영역을 포함하며 상기 제2이진화 영역보다 넓은 영역을 가지는 제2계수 연산영역을 설정하는 단계, 상기 오브젝트 인식시스템이 상기 제2계수 연산영역에 포함된 픽셀들의 픽셀 값들에 기초하여 제2이진화 계수를 결정하는 단계, 및 상기 오브젝트 인식시스템이 결정된 상기 제2이진화 계수를 통해 상기 제2이진화 영역에 대한 이진화를 수행하는 단계를 포함한다.

57. 发明专利

JP6260113B2 エッジの抽出方法および設備有权
公开(公告)号：JP6260113B2
公开(公告)日：2018-01-17
申请号：JP2013114204
申请日：2013-05-30
申请人：富士通株式会社
发明人：何源 , 孫俊 , 直井聡
IPC分类号： G06T7/13
CPC分类号： G06T5/20 , G06T5/003 , G06T7/12 , G06T7/181 , G06T2207/10008 , G06T2207/20192 , G06T2207/30176

58. 发明专利

JP6260112B2 画像処理装置、画像処理方法、スキャナーおよび記憶媒体有权
公开(公告)号：JP6260112B2
公开(公告)日：2018-01-17
申请号：JP2013114201
申请日：2013-05-30
申请人：富士通株式会社
发明人：潘攀 , 何源 , 孫俊 , 直井聡
IPC分类号： G06T7/181
CPC分类号： G06K9/00463 , G06K9/32 , G06K9/4638 , G06K2009/485 , G06K2209/01 , G06T5/002 , G06T5/40 , G06T7/12 , G06T7/181 , G06T2207/10008 , G06T2207/20192 , G06T2207/30176

59. 发明专利

JP6200506B2 試料の画像内の繊維および分枝構造の分析のための方法および装置有权
公开(公告)号：JP6200506B2
公开(公告)日：2017-09-20
申请号：JP2015530000
申请日：2013-08-28
申请人：モレキュラーデバイシーズ, エルエルシー
发明人：ザオ, ティン , ホン, ディフイ , コーエン, アブルムアイ.
IPC分类号： A61B5/00 , G06T7/60
CPC分类号： G06K9/00134 , G06K9/0014 , G06K9/00147 , G06T7/12 , G06T7/136 , G06T7/162 , G06T7/181 , G06T2207/10056 , G06T2207/30024

60. 发明专利

JP6091720B2 画像生成装置、画像生成方法及びコンピュータプログラム有权
标题翻译：图像生成装置，图像生成方法和计算机程序
公开(公告)号：JP6091720B2
公开(公告)日：2017-03-08
申请号：JP2016546281
申请日：2014-09-05
申请人：堺ディスプレイプロダクト株式会社
发明人：繁田光浩
IPC分类号： G08B25/00 , G06T1/00 , G06T3/00 , G06T3/60 , H04N7/18
CPC分类号： G06T3/0025 , G06T3/00 , G06T3/4007 , G06T3/403 , G06T3/60 , G06T5/006 , G06T5/50 , G06T7/11 , G06T7/181 , H04N7/181 , G06T2207/10016 , G06T2207/20164 , G06T2207/30232 , G06T2207/30236 , G06T2207/30256

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