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

JP2014106092A Three-dimensional position measurement device, three-dimensional position measurement device and three-dimensional position measurement program 有权
标题翻译：三维位置测量装置，三维位置测量装置和三维位置测量程序
公开(公告)号：JP2014106092A
公开(公告)日：2014-06-09
申请号：JP2012258811
申请日：2012-11-27
申请人： Fujitsu Ltd , 富士通株式会社
发明人： KANETAKE JUN , NAKAYAMA OSAFUMI
IPC分类号： G01B11/00 , B60R1/00 , B60R21/00 , G01C11/06 , G08G1/16
摘要： PROBLEM TO BE SOLVED: To suppress a rise in a process time upon calculating a three-dimensional position.SOLUTION: The three-dimensional position measurement device measuring a three-dimensional position of an object on the basis of a difference between positions of objects on images photographed by a camera at a mutually different time and a movement amount of the camera includes: an image acquisition part that sequentially acquires an image from the camera; a characteristic point extraction part that extracts a characteristic point from an image of a process object; a characteristic point collation part that collates a registration characteristic point with an extraction characteristic point serving as a characteristic point extracted from the image of the process object; a three-dimensional position determination part that calculates a three-dimensional position of the extraction characteristic point corresponding to the registration characteristic point and determining the three-dimensional position of the extraction characteristic point; and a characteristic point registration part that registers an extraction characteristic point excluding the characteristic point having the three-dimensional position determined among the extraction characteristic points as a registration characteristic point to be used for collating the registration characteristic point with a characteristic point to be extracted from an image of a next process object.
摘要翻译：要解决的问题：在计算三维位置时抑制处理时间的上升。解决方案：三维位置测量装置基于物体的位置之间的差异来测量物体的三维位置摄像机在相互不同的时间拍摄的图像和照相机的移动量包括：图像获取部分，其从相机顺序地获取图像; 特征点提取部，从处理对象的图像中提取特征点; 特征点对照部，其将登记特征点与用作从所述处理对象的图像提取的特征点的提取特征点进行整理; 三维位置确定部，其计算与登记特征点对应的提取特征点的三维位置，并确定提取特征点的三维位置; 以及特征点登记部，其将从提取特征点之间确定的具有所述三维位置的所述特征点以外的提取特征点登记为用于将所述登记特征点与要从所述提取特征点中提取的特征点进行对照的登记特征点下一个进程对象的映像。

2. 发明专利

JP2014044093A Speed estimation device, and speed estimation method 有权
标题翻译：速度估计装置和速度估计方法
公开(公告)号：JP2014044093A
公开(公告)日：2014-03-13
申请号：JP2012185864
申请日：2012-08-24
申请人： Fujitsu Ltd , 富士通株式会社
发明人： NAGAMURA KAZUKI , KITAURA ASAKO , NAKAYAMA OSAFUMI
IPC分类号： G01P3/36
摘要： PROBLEM TO BE SOLVED: To provide a speed estimation device which, even when feature points extracted from a plurality of images captured by an on-vehicle camera are distributed unevenly, can improve accuracy of estimating the speed of a vehicle based on those images.SOLUTION: A speed estimation device 1, for each of a plurality of points around a vehicle 10, extracts a corresponding first feature point on a first image captured at a first point, and also, and extracts a corresponding second feature point on a second image captured at a second point. The speed estimation device 1 specifies a range which a traveling direction of a vehicle heading from the first point to the second point can take, based on acceleration in a direction perpendicular to the traveling direction of the vehicle 10 and also horizontal to a road surface, projects each feature point on the image of one of the first image and the second image to the another image according to the traveling direction of the vehicle estimated within the range, estimates the traveling direction of the vehicle in such a manner that an estimation value indicating an error of a position between each of the projected feature points and corresponding feature point on the another image becomes minimum, and estimates a speed of the vehicle based on the estimated traveling direction.
摘要翻译：要解决的问题：提供一种速度估计装置，即使当从车载摄像机拍摄的多个图像中提取的特征点分布不均匀时，也可以提高基于这些图像的车辆速度的估计精度。：速度估计装置1，对于车辆10周围的多个点中的每一个，提取在第一点拍摄的第一图像上的对应的第一特征点，并且提取在捕获的第二图像上的对应的第二特征点在第二点。速度估计装置1基于与车辆10的行进方向垂直的方向上的加速度，并且还与水平面相对应地指定从第一点到第二点的车辆的行进方向可以取得的范围，根据在该范围内估计出的车辆的行进方向，将第一图像和第二图像中的一个图像的图像上的每个特征点投影到另一图像，以如下方式估计车辆的行进方向：指示每个投影特征点和另一图像上的对应特征点之间的位置的误差变得最小，并且基于估计的行进方向来估计车辆的速度。

3. 发明专利

JP2014109819A Image processing apparatus, image processing method, 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：JP2014109819A
公开(公告)日：2014-06-12
申请号：JP2012262772
申请日：2012-11-30
申请人： Fujitsu Ltd , 富士通株式会社
发明人： KANETAKE JUN , NAKAYAMA OSAFUMI
IPC分类号： G06T7/20 , G01B11/24 , G06T1/00 , G06T7/60
CPC分类号： H04N13/0282 , G06T7/579 , G06T2207/10016 , G06T2207/20012 , G06T2207/30264
摘要： PROBLEM TO BE SOLVED: To generate a proper shape for each object.SOLUTION: An image processing apparatus includes: an image acquisition part which acquires an image from a camera installed in a moving body; a feature point extraction part which extracts feature points from the images; a collation part which collates the feature points extracted from the images captured at different times; a position calculation part which calculates a three-dimensional position on the basis of a position of the collated feature point and a moving amount of the moving body; an accuracy calculation part which calculates accuracy of indicating an error in the three-dimensional position; a distribution determination part which detects an object from the image, and sets a threshold for each object on the basis of an accuracy distribution of the feature point generated for each object; a selection part which selects a feature point having higher accuracy than the threshold, for each object; and a generation part which generates an object shape by use of the selected feature point.
摘要翻译：要解决的问题：为每个物体生成适当的形状。解决方案：一种图像处理装置，包括：图像获取部，其从安装在移动体中的相机获取图像; 特征点提取部分，从图像中提取特征点; 对照部，其对从不同时刻拍摄的图像提取的特征点进行整理; 位置计算部，其基于所述核对特征点的位置和所述移动体的移动量来计算三维位置; 精度计算部，其计算指示三维位置的误差的精度; 分发确定部分，其从图像中检测对象，并且基于为每个对象生成的特征点的精度分布来设置每个对象的阈值; 对于每个对象，选择具有比阈值更高的精度的特征点的选择部分; 以及通过使用所选择的特征点来生成对象形状的生成部。

4. 发明专利

JP2014081874A Image processing device and image processing method 有权
标题翻译：图像处理设备和图像处理方法
公开(公告)号：JP2014081874A
公开(公告)日：2014-05-08
申请号：JP2012230850
申请日：2012-10-18
申请人： Fujitsu Ltd , 富士通株式会社
发明人： NAKAYAMA OSAFUMI
IPC分类号： G06T7/20 , G08G1/04 , H04N7/18
CPC分类号： G06K9/00791 , G06T5/008 , G06T7/246 , G06T2207/10016 , G06T2207/30261
摘要： PROBLEM TO BE SOLVED: To appropriately estimate an amount of movement of a moving object when estimating the amount of movement of the moving object by using images captured by an imaging unit installed on the moving image.SOLUTION: An image processing device includes: an extraction unit which uses images captured by one or more imaging units installed on a moving image to superpose a first edge image extracted from an image at a first time and a second edge image extracted from an image at a second time after the first time and extracts overlapping edges; a removal unit which removes the edges extracted by the extraction unit from the first edge image and/or the second edge image; a collation unit which performs collation processing on the first edge image and the second edge image one or both of which the edges are removed from by the removal unit; and an estimation unit which uses an amount of deviation between the edge images collated by the collation unit to estimate the amount of movement of a moving object.
摘要翻译：要解决的问题：通过使用由安装在运动图像上的成像单元拍摄的图像来估计移动物体的移动量来适当估计移动物体的移动量。解决方案：图像处理装置包括：提取单元，其使用由安装在运动图像上的一个或多个成像单元捕获的图像来叠加从第一次起的图像提取的第一边缘图像，以及在第一时间之后的第二时间从图像提取的第二边缘图像，并且提取重叠边缘; 去除单元，其从所述第一边缘图像和/或所述第二边缘图像中去除由所述提取单元提取的边缘; 对照单元，对所述第一边缘图像和所述第二边缘图像执行对照处理，所述第一边缘图像中的一个或两个边缘被所述移除单元除去; 以及估计单元，其使用由对照单元对照的边缘图像之间的偏差量来估计移动物体的移动量。

5. 发明专利

JP2014240753A 距離測定装置、距離測定方法、およびプログラム有权
标题翻译：距离测量装置，距离测量方法和程序
公开(公告)号：JP2014240753A
公开(公告)日：2014-12-25
申请号：JP2013122315
申请日：2013-06-11
申请人：富士通株式会社 , Fujitsu Ltd
发明人： KANETAKE JUN , NAKAYAMA OSAFUMI
IPC分类号： G01B11/00 , G01C3/06 , G06T1/00
CPC分类号： H04N13/0221 , G01S1/00 , G01S5/16 , G01S11/12 , G06T7/579 , G06T2207/30252 , H04N2013/0081
摘要：【課題】物体までの距離を誤って測定することを防止する。【解決手段】三次元位置算出部は、車両の撮影装置により第1の位置で撮影された第1の画像、および第1の位置とは異なる第2の位置で撮影された第2の画像を受付ける。三次元位置算出部は、車両の移動量検出装置から、第1の位置と第2の位置との間の車両の移動量を受付け、第1の画像、第2の画像、および移動量に基づき、周囲の物体の三次元位置を算出する。判別部は、算出された三次元位置に基づき設定された第3の位置を、第1の画像および第2の画像のうち一方の画像に投影した第4の位置と、一方の画像における物体の位置とのずれが所定値以上の場合に、算出された三次元位置が誤りであると判別する。出力部は、算出された三次元位置が、誤りであると判別されなかった場合には距離測定対象とし、誤りであると判別された場合に距離測定対象から除外する。【選択図】図1
摘要翻译：要解决的问题：防止对物体的距离的错误测量。解决方案：三维位置计算单元接收由车辆成像装置在第一位置拾取的第一图像和在第二位置拾取的第二图像不同于第一个位置。三维位置计算单元从车辆的移动量检测器接收第一位置和第二位置之间的车辆的移动量，并且基于第一图像计算周围物体的三维位置，第二图像和移动量。如果通过将基于所计算的三维位置设置的第三位置投影到第一图像和第二图像之一而获得的第四位置与对象位置之间的偏差，则识别单元判别计算出的三维位置是误差在第一和第二图像之一中不小于预定值。如果计算出的三维位置未被识别为误差，则输出单元将计算出的三维位置确定为距离测量对象，并且如果三维位置被区分为距离测量目标，则将三维位置从距离测量目标中排除错误。

6. 发明专利

JP2014109870A Operation evaluation device, method and program, and operation-evaluating on-vehicle machine 有权
标题翻译：操作评估装置，方法与程序以及操作评估机动车
公开(公告)号：JP2014109870A
公开(公告)日：2014-06-12
申请号：JP2012263313
申请日：2012-11-30
申请人： Fujitsu Ltd , 富士通株式会社
发明人： KITAURA ASAKO , NAKAYAMA OSAFUMI
IPC分类号： G08G1/16 , B60W40/06
摘要： PROBLEM TO BE SOLVED: To improve accuracy of an operation evaluation by accurately evaluating a degree of risk at an intersection.SOLUTION: An intersection entrance start position estimation part 18 estimates an entrance start position of a vehicle at an evaluation-target intersection on the basis of on-vehicle data obtained by an on-vehicle device incorporated in the vehicle; a visibility-enabled position estimation part 20 identifies and compares a required visibility zone at the evaluation-target intersection on the basis of the on-vehicle data and determines individually a range that is visible to a crew of the vehicle, of the identified required visibility zone, at individual positions forward and backward of the entrance start position, of the evaluation-target intersection, thus estimating the visibility-enabled position of the vehicle at the evaluation-target intersection; and an intersection risk-degree evaluation part 22 evaluates a degree of risk at the evaluation-target intersection on the basis of at least positional relationship between the entrance start position of the vehicle at the evaluation-target intersection and the visibility-enabled position of the vehicle at the evaluation-target intersection.
摘要翻译：要解决的问题：通过精确评估十字路口处的风险程度，提高作业评价的准确性。交叉口入口开始位置估计部18基于评估对象交叉点估计车辆的入口开始位置由车载装置获得的车载数据; 可见度启用位置估计部件20基于车载数据识别并比较评估目标交叉口处的所需可见度区域，并且单独确定车辆的乘员人员可见的范围所识别的所需可见度在评价对象交叉点的入口开始位置的前后的各个位置，从而估计车辆在评价对象交叉路口处的可见度能力位置; 并且交叉路口风险度评价部22基于评价对象交叉口处的车辆的入口开始位置与评价对象交叉路口的可见度开启位置之间的至少位置关系来评估评价对象交叉路口处的风险程度车辆在评估目标交叉路口。

7. 发明专利

JP2014106147A Intersection detection method and intersection detection system 有权
标题翻译：交叉检测方法和交叉检测系统
公开(公告)号：JP2014106147A
公开(公告)日：2014-06-09
申请号：JP2012259860
申请日：2012-11-28
申请人： Fujitsu Ltd , 富士通株式会社
发明人： KITAURA ASAKO , KARASUYA AYU , NAKAYAMA OSAFUMI
IPC分类号： G01C21/28 , G08G1/16
摘要： PROBLEM TO BE SOLVED: To accurately detect a start position of an intersection.SOLUTION: An intersection detection method includes: acquiring in-vehicle data including first data to be used in the case of examining a state around a traveling object and first position information indicating a position of the traveling object and map data including second position information indicating the position of an intersection; selecting the in-vehicle data acquired at a spot within a predetermined distance from the intersection on the basis of the first position information and the second position information; determining whether or not the characteristics of the intersection exist in the first data of the selected in-vehicle data; specifying the in-vehicle data acquired within the intersection on the basis of the determination result; determining the start position of a search range on the basis of the specified in-vehicle data; determining the search direction of the in-vehicle data as a time direction which is reverse to the recording time direction of the in-vehicle data; successively selecting the in-vehicle data to be determined in accordance with the search direction; determining whether or not characteristics corresponding to an intersection start position appear in the in-vehicle data to be determined; and specifying the in-vehicle data of the intersection start position.
摘要翻译：要解决的问题：准确地检测十字路口的起始位置。交叉口检测方法包括：获取包括在检查行驶物体周围的状态的情况下使用的第一数据的车内数据和指示行进对象的位置和包括表示交叉路口位置的第二位置信息的地图数据; 基于所述第一位置信息和所述第二位置信息，选择在距离所述交点的预定距离内的点处获取的车载数据; 确定所选择的车内数据的第一数据中是否存在交集的特征; 基于所述确定结果指定在所述十字路口内获取的车内数据; 基于指定的车内数据确定搜索范围的开始位置; 确定所述车载数据的搜索方向为与所述车载数据的记录时间方向相反的时间方向; 根据搜索方向依次选择要确定的车内数据; 确定对应于交叉点开始位置的特性是否出现在待确定的车内数据中; 并指定交叉口开始位置的车内数据。

8. 发明专利

JP2014194361A Movement distance estimation device, movement distance estimation method, and program 有权
标题翻译：运动距离估计装置，运动距离估计方法和程序
公开(公告)号：JP2014194361A
公开(公告)日：2014-10-09
申请号：JP2013070278
申请日：2013-03-28
申请人： Fujitsu Ltd , 富士通株式会社
发明人： NAGAMURA KAZUKI , KITAURA ASAKO , NAKAYAMA OSAFUMI
IPC分类号： G01B11/00 , G01C7/04 , G01P3/36 , G06T7/20
CPC分类号： G07C5/00 , G06K9/00208 , G06K9/00798 , G06T7/248 , G06T2207/30236
摘要： PROBLEM TO BE SOLVED: To enable a movement distance to be estimated based on a video image captured from a movable body moving on a road surface even when a road surface pattern is poor.SOLUTION: A relative movement processing part 312 obtains an amount of a relative movement of a movable body between a first time and a second time on the basis of a first image and a second image included in video image information inside a storage part 311. A first scale calculation part 313 obtains a first scale of the amount of the relative movement on the basis of a partial image in a road surface region in a third image included in the video image information. A second scale calculation part 314 obtains a second scale of the amount of the relative movement on the basis of a condition that a space point in a three-dimensional space corresponding to a characteristic point in a region other than the road surface region exists on a road surface. A movement distance calculation part 315 obtains a movement distance between the first time and the second time on the basis of the amount of the relative movement and the first scale or the second scale.
摘要翻译：要解决的问题：即使当路面图案差时，也可以基于从在路面上移动的移动体捕获的视频图像来估计移动距离。解决方案：相对运动处理部分312获得基于包含在存储部311中的视频图像信息中的第一图像和第二图像，可移动体在第一时间和第二时间之间的相对移动。第一刻度计算部313获得第一刻度基于包括在视频图像信息中的第三图像中的路面区域中的部分图像的相对运动。第二比例计算部314基于与在路面区域以外的区域中的特征点对应的三维空间中的空间点存在的条件，获得相对移动量的第二比例路面。移动距离计算部315基于相对移动量和第一刻度或第二刻度来获得第一时刻与第二时刻之间的移动距离。

9. 发明专利

JP2014109875A Intersection detecting method and intersection detecting system 有权
标题翻译：交叉检测方法和接口检测系统
公开(公告)号：JP2014109875A
公开(公告)日：2014-06-12
申请号：JP2012263354
申请日：2012-11-30
申请人： Fujitsu Ltd , 富士通株式会社
发明人： KARASUYA AYU , KITAURA ASAKO , NAKAYAMA OSAFUMI
IPC分类号： G06T1/00 , G08G1/16 , H04N5/225
摘要： PROBLEM TO BE SOLVED: To accurately detect an intersection entry port.SOLUTION: In an intersection detecting method for detecting an intersection entry port by using an image captured by a camera mounted on a mobile body, an image is acquired, a feature quantity showing likelihood of an intersection in the image is calculated, it is determined whether or not the image is an image captured in an intersection on the basis of the feature quantity, the image captured in an intersection is determined as a start image for search, images are sequentially selected in reverse order of photographic order of the images, a level difference of a motion vector of the image is searched for and an intersection entry port is identified based on the level difference of the detected motion vector.
摘要翻译：要解决的问题：准确地检测交叉口入口。解决方案：在通过使用由安装在移动体上的相机拍摄的图像来检测交叉口入口的交叉点检测方法中，获取图像，显示可能性的特征量计算图像中的交叉点，根据特征量确定图像是否是在交叉路口中拍摄的图像，将在交叉点中捕获的图像确定为用于搜索的开始图像，图像依次以图像的照相顺序的相反顺序选择，搜索图像的运动矢量的电平差，并且基于检测到的运动矢量的电平差来识别交叉入口端口。

10. 发明专利

JP2011242198A Position specification device, position specification method, and program 有权
标题翻译：位置规范设备，位置规范方法和程序
公开(公告)号：JP2011242198A
公开(公告)日：2011-12-01
申请号：JP2010113007
申请日：2010-05-17
申请人： Fujitsu Ltd , 富士通株式会社
发明人： OKADA YASUTAKA , NAKAYAMA OSAFUMI
IPC分类号： G01B11/00 , G08G1/16
摘要： PROBLEM TO BE SOLVED: To allow identification of a position of an object, which is performed by using a monocular camera, to be achieved in a wide range.SOLUTION: A first position identification unit 7a identifies a position of a first object, based on two pieces of first picked-up images concerning the first object, which are obtained by respectively performing imaging from two known points whose positions are known. A second position identification unit 7b identifies a position of a second object. The identification of the position of the second object is performed based on the position of the first object identified by the first position identification unit 7a, one piece of the second picked-up image concerning the second object, including an image of the first object, and the image including the image of the second object between the first picked-up images. The second picked-up image is the image obtained by performing imaging from an arbitrary position on a route where the two known points are connected and a route locus is known.
摘要翻译：要解决的问题：允许通过使用单目相机进行的物体的位置的识别在宽范围内实现。解决方案：第一位置识别单元7a基于通过分别从已知位置的两个已知点分别执行成像而获得的关于第一对象的两个第一拍摄图像来识别第一对象的位置。第二位置识别单元7b识别第二物体的位置。基于由第一位置识别单元7a识别的第一对象的位置执行第二对象的位置的识别，包括第二对象的第二拍摄图像，包括第一对象的图像，并且所述图像包括所述第一拾取图像之间的所述第二对象的图像。第二拍摄图像是通过连接两个已知点的路线上的任意位置进行成像而获得的图像，并且已知路线轨迹。版权所有（C）2012，JPO＆INPIT

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