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. 发明专利

JP2014112799A Camera system 有权
标题翻译：相机系统
公开(公告)号：JP2014112799A
公开(公告)日：2014-06-19
申请号：JP2012266576
申请日：2012-12-05
申请人： Secom Co Ltd , セコム株式会社
发明人： TAKAHASHI ISAO , KUROKAWA TAKAHARU
IPC分类号： H04N5/225 , G01C11/04 , G06T1/00 , H04N5/232
摘要： PROBLEM TO BE SOLVED: To provide a camera system that enables efficient work of camera calibration and irradiation point setup for three-dimensional measurements.SOLUTION: A camera system includes a control device 4 for processing an image imaged by one or a plurality of cameras 2. The control device 4 includes: a storage unit 41 for storing at least a plurality of irradiation request positions in a monitoring space required to form an irradiation point; a recommended light source position calculation unit 423 for determining a position that produces a minimum sum of distances from the plurality of irradiation request positions to be a recommended light source position when irradiating the plurality of irradiation request positions; and a display communication unit 43 for outputting the recommended light source position.
摘要翻译：要解决的问题：提供能够有效地进行相机校准和用于三维测量的照射点设置的相机系统。解决方案：相机系统包括用于处理由一个或多个照相机2成像的图像的控制装置4 控制装置4包括：存储单元41，用于在形成照射点所需的监视空间中存储至少多个照射请求位置; 推荐的光源位置计算单元423，用于当照射多个照射请求位置时，确定从多个照射请求位置产生最小距离的和作为推荐光源位置的位置; 以及用于输出推荐光源位置的显示通信单元43。

2. 发明专利

JP2013191143A Image processing device 有权
标题翻译：图像处理装置
公开(公告)号：JP2013191143A
公开(公告)日：2013-09-26
申请号：JP2012058408
申请日：2012-03-15
申请人： Secom Co Ltd , セコム株式会社
发明人： MUNEKATA TAKUMI , KUROKAWA TAKAHARU
IPC分类号： G06T7/00 , G06T1/00
摘要： PROBLEM TO BE SOLVED: To classify discrimination targets having a common feature into subdivisions while securing discrimination accuracy using a cascade-type learning discriminator.SOLUTION: An image processing device comprises: an image acquisition unit 102 that acquires an image; and target image discriminators 112, 113, and 114 that have learned discrimination targets of respective classes. The target image discriminators 112, 113, and 114 are connected in series so as to be higher in order as an error rate is lower, and discriminate a discrimination target from the highest order to classify it into each class.
摘要翻译：要解决的问题：将具有共同特征的识别目标分类为细分，同时使用级联式学习鉴别器确保辨别精度。解决方案：图像处理装置包括：图像获取单元102，其获取图像; 以及已经学习各种类别的识别目标的目标图像鉴别器112,113和114。目标图像识别器112,113和114串联连接，以便按照错误率较低的顺序更高，并且将辨别目标与最高级别进行区分以将其分类成每个类别。

3. 发明专利

JP2013143069A Image processing apparatus 有权
标题翻译：图像处理设备
公开(公告)号：JP2013143069A
公开(公告)日：2013-07-22
申请号：JP2012003747
申请日：2012-01-12
申请人： Secom Co Ltd , セコム株式会社
发明人： SAGAWA TATSUYUKI , KUROKAWA TAKAHARU
IPC分类号： G06T7/20 , H04N7/18
摘要： PROBLEM TO BE SOLVED: To solve such a problem that, when a person moves an installed object in a monitoring space, a change area integrated with the installed object is generated on a monitoring image and, if the change area is uniformly processed for protecting privacy, it is hard to visually recognize the shapes of the person and a dolly.SOLUTION: An image processing apparatus causes a storage unit 3 to store a background image 30 including an installed object. A change area extraction unit 40 obtains a change area between the background image 30 and an input image. An installation change detecting unit 42 collates the background image 30 with the input image to obtain an installed object-derived area generated by moving the installed object in the change area. A mask image generation unit 44 excludes the installed object-derived area from an object of a masking process when performing a masking process for the input image in the change area.
摘要翻译：要解决的问题：为了解决这样一个问题，即当人们将监控空间中安装的物体移动时，在监视图像上产生与安装对象集成的变化区域，并且如果改变区域被均匀地处理以保护隐私，难以在视觉上识别个人和个人的形状。解决方案：图像处理装置使存储单元3存储包括安装对象的背景图像30。改变区域提取单元40获得背景图像30和输入图像之间的变化区域。安装变化检测单元42将背景图像30与输入图像进行核对，以获得通过在安装区域中移动安装的对象而产生的安装的对象导出区域。掩模图像生成单元44在对变化区域中的输入图像执行掩蔽处理时，从掩蔽处理对象中排除所安装的对象导出区域。

4. 发明专利

JP2010049297A Image monitoring device 有权
标题翻译：图像监控装置
公开(公告)号：JP2010049297A
公开(公告)日：2010-03-04
申请号：JP2008210280
申请日：2008-08-19
申请人： Secom Co Ltd , セコム株式会社
发明人： KUROKAWA TAKAHARU , TAKAI FUMIHIKO
IPC分类号： G06T7/20
摘要： PROBLEM TO BE SOLVED: To solve the problem that in an image monitoring device for specifying a target object appearing in a monitoring image, the state of occlusion is changed between components of the target object due to the movement of the target object, and the accuracy of identifying the target object based on comparison with a previous image is reduced. SOLUTION: An image (reference image 401) of a component of an object to be compared in identifying the image of the target object from the monitoring image is stored together with a reference image occlusion degree 402 into a storage part 4. An object model 400 is arranged at a candidate position in the monitoring image, and the extracted image is compared with the reference image 401. When similarity equal to or more than a reference value is obtained for image features, it is determined that the target object exists at the candidate position. In this case, for the component where the degree of occlusion of the partial image for each component at the candidate position is equal to or less than the reference image occlusion degree 402, the reference image 401 and the reference image occlusion degree 402 are updated with the current partial image and the degree of occlusion. COPYRIGHT: (C)2010,JPO&INPIT
摘要翻译：要解决的问题为了解决在用于指定出现在监视图像中的目标对象的图像监视装置中的问题，由于目标对象的移动而在目标对象的各部件之间改变遮挡状态，并且基于与先前图像的比较来识别目标对象的准确性被降低。解决方案：将来自监视图像的目标物体的图像识别中要比较的对象的分量的图像（参考图像401）与参考图像遮挡度402一起存储到存储部分4中。对象模型400被布置在监视图像中的候选位置处，并且将提取的图像与参考图像401进行比较。当对于图像特征获得等于或大于参考值的相似度时，确定目标对象存在在候选人位置。在这种情况下，对于候选位置处的每个分量的部分图像的遮挡程度等于或小于参考图像遮挡度402的分量，参考图像401和参考图像遮挡度402被更新为当前部分图像和遮挡度。版权所有（C）2010，JPO＆INPIT

5. 发明专利

JP2014149597A Passing object detection device 有权
标题翻译：通过物体检测装置
公开(公告)号：JP2014149597A
公开(公告)日：2014-08-21
申请号：JP2013016883
申请日：2013-01-31
申请人： Secom Co Ltd , セコム株式会社
发明人： MAEDA MASAHIRO , NAKAZAWA YUJI , OSHIMA NAOYA , KUROKAWA TAKAHARU , UJIIE HIDENORI
IPC分类号： G06T7/20 , H04N7/18
摘要： PROBLEM TO BE SOLVED: To keep detecting passing objects regardless of whether a door is open or closed.SOLUTION: A passing object detection device preliminarily stores a door shape pattern including a shape feature of an opening generated by opening a door according to the door opening level, calculates the similarity by comparing a photographic image with the shape pattern, divides the photographic image into a door area and an outer door area matching it with a shape pattern which has the most similar door opening level with that of the photographic image, stores background information of a space excluding a passing object in each area, calculates the difference between each pixel of the photographic image in each area and the background information of the area, extracts a group of pixels having the difference larger than a prescribed reference as a passing object area, stores distribution data of pixel values of the door area as background information in the door area, and stores a pixel value of each pixel in the outer door area as background information in the outer door area.
摘要翻译：要解决的问题：无论门是打开还是关闭，都不断检测通过的物体。解决方案：通过物体检测装置预先存储包括根据门开口产生的打开门的开口的形状特征的门形图案通过将摄影图像与形状图案进行比较来计算相似度，将摄影图像分成与区域相匹配的门区域和外门区域，其门形状图案与摄影图像的门开口水平最相似，存储计算每个区域中除了通过对象之外的空间的背景信息，计算每个区域中的摄影图像的每个像素与该区域的背景信息之间的差异，提取具有大于规定参考值的差异的一组像素作为通过对象区域，将门区域的像素值的分布数据作为背景信息存储在门区域中，并存储a 外门区域中的每个像素的像素值作为外门区域中的背景信息。

6. 发明专利

JP2014010717A Area division device 有权
标题翻译：区域设备
公开(公告)号：JP2014010717A
公开(公告)日：2014-01-20
申请号：JP2012147990
申请日：2012-06-29
申请人： Secom Co Ltd , セコム株式会社
发明人： LI YEQIU , KUROKAWA TAKAHARU
IPC分类号： G06T7/60
摘要： PROBLEM TO BE SOLVED: To provide an area division device for dividing an image into an object region and a background region based on plural kinds of image feature amounts capable of increasing the extraction accuracy in an object region even when causes occur to reduce the accuracy varying depending on the part of the object.SOLUTION: In a tentative setting of a belonging region in which each pixel belongs and one kind of feature to be evaluated for evaluating the pixel in the plural kinds of image features, an energy calculation part 412 calculates an integrated evaluation value by summing, in the image, the degree of belongingness representing the degree of plausibility that each pixel belongs in the belonging region in which features to be evaluated is tentatively determined in the setting. A region determination section 413 creates plural tentative settings and compares with the integrated evaluation value made by the tentative setting calculated by the energy calculation part 412, and determines a belonging region where the plausibility is the maximum in the entire image in the tentative settings as a region division result.
摘要翻译：要解决的问题：提供一种区域划分装置，用于基于能够增加对象区域中的提取精度的多种图像特征量，将图像分割成对象区域和背景区域，即使当发生原因而降低精度变化取决于对象的一部分。解决方案：能量计算部分412在每个像素所属的属性区域的暂定设置和用于评估多种图像特征中的像素的评估特征中，在该图像中，通过在图像中求和表示每个像素属于其中要评估的特征的属性区域中的似然度的归属度在该图像中求和来确定综合评价值。区域确定部分413创建多个临时设置并与由能量计算部分412计算出的暂定设置所得到的综合评价值进行比较，并将暂定设置中的整个图像中的合理性最大值的归属区域确定为地区划分结果。

7. 发明专利

JP2012208850A Abnormality detection apparatus 有权
标题翻译：异常检测装置
公开(公告)号：JP2012208850A
公开(公告)日：2012-10-25
申请号：JP2011075396
申请日：2011-03-30
申请人： Secom Co Ltd , セコム株式会社
发明人： OSHIMA NAOYA , KUROKAWA TAKAHARU
IPC分类号： G08B13/196 , G06T1/00 , G08B21/02 , G08B25/00
摘要： PROBLEM TO BE SOLVED: To provide an abnormality detection apparatus capable of automatically detecting an abnormal condition and performing an emergency notification even when a plurality of persons are controlled to hamper a notification operation.SOLUTION: The abnormality detection apparatus comprises: person area extraction means 220 for extracting a person as a person area from a monitoring image; posture determination means 222 for determining a posture of the person from the person area; person tracking means 221 for tracking the person area of the same person in the monitoring image sequentially acquired; posture change determination means 223 for determining whether or not there is a posture change for each of the tracked persons; and abnormality determination means 225 for determining that there is the abnormality when the number of persons having the posture change within a predetermined time determined as almost the same is equal to or greater than a reference value.
摘要翻译：要解决的问题：即使当控制多个人妨碍通知操作时，提供能够自动检测异常状况并执行紧急通知的异常检测装置。解决方案：异常检测装置包括：人区域提取装置220，用于从监视图像中提取人为人区域; 姿势确定装置222，用于从人区域确定人的姿势; 个人跟踪装置221，用于跟踪顺序取得的监视图像中的同一人的人物区域; 姿势变化确定装置223，用于确定每个跟踪人员是否存在姿势变化; 以及异常判定单元225，用于当在几乎相同的预定时间内具有姿势改变的人数等于或大于参考值时，确定存在异常。版权所有（C）2013，JPO＆INPIT

8. 发明专利

JP2010277431A Object detection device 有权
标题翻译：对象检测装置
公开(公告)号：JP2010277431A
公开(公告)日：2010-12-09
申请号：JP2009130922
申请日：2009-05-29
申请人： Secom Co Ltd , セコム株式会社
发明人： KUROKAWA TAKAHARU
IPC分类号： G06T7/00 , G08B13/196 , H04N7/18
摘要： PROBLEM TO BE SOLVED: To improve accuracy of detection in an object detection device for detecting the image of an object appearing in a monitoring image.
SOLUTION: An area setting means 40 sets the candidate area of the image of an object in a monitoring image. Alternatively, an image division means 41 performs segmentation processing to the monitoring image with a plurality of division granularity, and divides the monitoring image into a plurality of image fragments from pixels whose pixel values have prescribed similarity for each division granularity. A degree of assignment calculation means 43 calculates the degree of object assignment increasing according to the number of pixels inside a candidate area about the image fragments and the degree of background assignment increasing according to the number of pixels outside the candidate area. A degree of assignment integration means 44 calculates an integral area evaluation value by integrating the plurality of degree of object assignment or the degree of background assignment calculated about the different division granularity. On the basis of the integral area evaluation value, the existence or candidate area of the detection object corresponding to the candidate area is corrected.
COPYRIGHT: (C)2011,JPO&INPIT
摘要翻译：要解决的问题：提高用于检测出现在监视图像中的对象的图像的物体检测装置中的检测精度。
解决方案：区域设置装置40在监视图像中设置对象的图像的候选区域。或者，图像分割装置41以多个分割粒度对监视图像进行分割处理，并且将每个分割粒度的像素值具有规定的相似度的像素的监视图像分割为多个图像片段。分配度计算装置43根据候选区域内关于图像片段的像素数目和根据候选区域外的像素数量增加背景分配的程度来增加对象分配的程度。分配度整合装置44通过对多个对象分配度进行积分或者关于不同分割粒度计算的背景分配度来计算积分面积评价值。基于积分面积评估值，校正与候选区域对应的检测对象的存在或候选区域。版权所有（C）2011，JPO＆INPIT

9. 发明专利

JP2010277296A Object detection device 有权
标题翻译：对象检测装置
公开(公告)号：JP2010277296A
公开(公告)日：2010-12-09
申请号：JP2009128519
申请日：2009-05-28
申请人： Secom Co Ltd , セコム株式会社
发明人： KUROKAWA TAKAHARU
IPC分类号： G06T7/20 , G06T1/00 , G06T7/00 , H04N7/18
摘要： PROBLEM TO BE SOLVED: To improve accuracy of detection in an object detection device for detecting the image of an object appearing in a monitoring image.
SOLUTION: A feature area extraction means 40 extracts a feature area having the features of an object from a monitoring image. Alternatively, an image division means 41 performs segmentation processing to the monitoring image, and divides the monitoring image into a plurality of image fragments configured of pixels whose pixel values have prescribed similarities, respectively. An existence determination means 43 detects any uncertain fragment striding the inside/outside of the feature area, and calculates an inside evaluation value increasing according to the number of pixels inside the feature area among the uncertain fragments, and determines whether any object exists in the feature value by using the inside evaluation value as the extraction error of the feature area.
COPYRIGHT: (C)2011,JPO&INPIT
摘要翻译：要解决的问题：提高用于检测出现在监视图像中的对象的图像的物体检测装置中的检测精度。解决方案：特征区域提取装置40从监视图像中提取具有对象特征的特征区域。或者，图像分割装置41对监视图像进行分割处理，并将监视图像分割为由像素值具有规定的相似度的像素构成的多个图像片段。存在确定装置43检测跨越特征区域的内部/外部的任何不确定的片段，并且根据不确定片段内的特征区域内的像素数量来计算内部评估值增加，并且确定特征中是否存在任何对象通过使用内部评价值作为特征区域的提取误差来计算。版权所有（C）2011，JPO＆INPIT

10. 发明专利

JP2014127131A Background information generation device and object detection device 有权
标题翻译：背景技术生成装置和对象检测装置
公开(公告)号：JP2014127131A
公开(公告)日：2014-07-07
申请号：JP2012285237
申请日：2012-12-27
申请人： Secom Co Ltd , セコム株式会社
发明人： OSHIMA NAOYA , KUROKAWA TAKAHARU
IPC分类号： G08B25/00 , G06T7/20
摘要： PROBLEM TO BE SOLVED: To continuously detect an object such as a person etc., without reference to whether a shield is displaced.SOLUTION: A background information generation device includes: a first photography part 2a which photographs a monitor space from a first view point; a second photography part 2b which photographs a shield space, which is not photographed from the first view point when a door is at a stationary position in the monitor space and photographed from the first view point when the door is displaced from the stationary position to the first view point, from a second view point where the shield space can be photographed at least when the door is at the stationary position; a door state detection part 320 which detects the door being displaced; and a background information generation part 321 which generates background information at the first view point from a first image photographed from the first view point when the door state detection part 320 does not detect the displacement yet, and generates background information at the first view point by using the background information generated from a second image that the second photography part 2b photographs when the door state detection part 320 detects the displacement.
摘要翻译：要解决的问题：连续地检测诸如人等的物体，而不考虑屏蔽是否被移位。解决方案：背景信息产生装置包括：第一摄影部分2a，从第一视点拍摄监视器空间 ; 第二拍摄部分2b，当门在静止位置时在监视器空间中从第一视点拍摄并且当门从静止位置移动到第一视点时从第一视点拍摄时不拍摄屏蔽空间从第二视点开始，至少当门处于静止位置时，能够拍摄屏蔽空间的第二视点; 检测门被移位的门状态检测部320; 以及背景信息生成部321，其在门状态检测部320还未检测到位移时从从第一观察点拍摄的第一图像在第一视点生成背景信息，并且在第一视点由使用从第二图像生成的背景信息，第二摄影部分2b在门状态检测部件320检测到位移时拍摄。

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