公开(公告)号：EP2840528A2

公开(公告)日：2015-02-25

申请号：EP14181202.4

申请日：2014-08-15

申请人： Ricoh Company, Ltd.

发明人： You, Ganmei ,  Lu, Yaojie ,  Shi, Zhongchao ,  Wang, Gang ,  Liu, Tong

IPC分类号： G06K9/00

CPC分类号： G06K9/00791 ,  G06T7/277

摘要： A method and an apparatus for tracking an object are disclosed. The method includes the steps of detecting, from a disparity map of a current frame, a position of the object including a size of the object and a center of the object; predicting the position of the object in the disparity map of the current frame based on the positions of the object detected from the disparity maps of a predetermined number of frames immediately before the current frame; and correcting the position of the object detected from the disparity map of the current frame to obtain a corrected position based on a confidence level of the size of the object detected from the disparity map of the previous frame, if the position of the object detected from the disparity map of the current frame matches the predicted position of the object.

摘要翻译： 公开了一种跟踪对象的方法和装置。 该方法包括以下步骤：从当前帧的视差图检测包括对象的大小和对象的中心的对象的位置; 基于从当前帧之前的预定数量帧的视差图中检测到的对象的位置来预测对象在当前帧的视差图中的位置; 以及根据从当前帧的视差图检测到的对象的位置，根据从前一帧的视差图检测到的对象的大小的置信度来校正从当前帧的视差图检测到的对象的位置，如果从 当前帧的视差图匹配对象的预测位置。

公开(公告)号：EP2720193A2

公开(公告)日：2014-04-16

申请号：EP13181638.1

申请日：2013-08-26

申请人： Ricoh Company, Ltd.

发明人： You, Ganmei ,  Lu, Yaojie ,  Shi, Zhongchao ,  Wang, Gang

IPC分类号： G06T7/00 ,  G08G1/00 ,  G06T7/20

CPC分类号： G06T7/70 ,  G06T2207/10021 ,  G06T2207/30252 ,  G06T2207/30256

摘要： Disclosed are a method and a system for detecting an uneven road surface. The method comprises a step of adopting two depth cameras to capture a depth map, and calculating a disparity map; a step of obtaining a V-disparity map by adopting a V-disparity algorithm based on the disparity map; a step of finding an oblique line corresponding to the road surface so as to let a direction of the oblique line be a direction of the road surface; a step of calculating a projection of cross sectional surface of the road surface based on the disparity map and coordinates of disappearance points; and a step of fitting the projection of the cross sectional surface of the road surface to obtain a road line on the cross sectional surface so as to extract the road surface, and letting points located below the fitted road line be road points.

摘要翻译： 公开了一种用于检测不平整路面的方法和系统。 该方法包括采用两个深度相机拍摄深度图并计算视差图的步骤; 通过采用基于视差图的V视差算法获得V视差图的步骤; 找出与路面对应的倾斜线，使倾斜线的方向成为路面的方向的步骤; 基于视差图和消失点坐标计算路面横截面投影的步骤; 以及将路面的横截面的突起嵌合以在截面上得到路线以提取路面的步骤，使位于装配的路线下方的点为道路点。

公开(公告)号：EP2728546B1

公开(公告)日：2020-02-12

申请号：EP12192685.1

申请日：2012-11-14

申请人： Ricoh Company, Ltd.

发明人： You, Ganmei ,  Lu, Yaojie ,  Zheng, Jichuan

IPC分类号： G06T7/73 ,  G06K9/00

公开(公告)号：EP2840528A3

公开(公告)日：2015-03-25

申请号：EP14181202.4

申请日：2014-08-15

申请人： Ricoh Company, Ltd.

发明人： You, Ganmei ,  Lu, Yaojie ,  Shi, Zhongchao ,  Wang, Gang ,  Liu, Tong

IPC分类号： G06K9/00 ,  G06T7/20

CPC分类号： G06K9/00791 ,  G06T7/277

摘要： A method and an apparatus for tracking an object are disclosed. The method includes the steps of detecting, from a disparity map of a current frame, a position of the object including a size of the object and a center of the object; predicting the position of the object in the disparity map of the current frame based on the positions of the object detected from the disparity maps of a predetermined number of frames immediately before the current frame; and correcting the position of the object detected from the disparity map of the current frame to obtain a corrected position based on a confidence level of the size of the object detected from the disparity map of the previous frame, if the position of the object detected from the disparity map of the current frame matches the predicted position of the object.

摘要翻译： 公开了一种用于追踪对象的方法和设备。 该方法包括以下步骤：从当前帧的视差图中检测包括对象的大小和对象的中心的对象的位置; 基于紧接在当前帧之前的预定数量的帧的视差图中检测到的对象的位置，预测对象在当前帧的视差图中的位置; 并根据从前一帧的视差图中检测到的物体的大小的置信度来校正从当前帧的视差图中检测到的物体的位置，以获得校正后的位置， 当前帧的视差图匹配对象的预测位置。

公开(公告)号：EP2608111A3

公开(公告)日：2014-09-17

申请号：EP12196446.4

申请日：2012-12-11

申请人： Ricoh Company, Ltd.

发明人： You, Ganmei ,  Zheng, Jichuan

IPC分类号： G06K9/00 ,  G06T7/00

CPC分类号： G06K9/00798 ,  G01C11/04 ,  G01C21/28 ,  G06K9/00208 ,  G06K9/4609 ,  G06T7/13 ,  G06T2207/10028 ,  G06T2207/30256

摘要： Disclosed are a road line detection method and a road line detection device. The road line detection method comprises a step of obtaining a first disparity map including one or more road regions and a corresponding V-disparity image; a step of sequentially detecting plural sloped line segments in the corresponding V-disparity image according to a big-to-small order of disparities and a big-to-small order of V-values, to serve as plural sequentially adjacent road surfaces; a step of obtaining a second disparity map of plural road line regions of interest corresponding to the plural sloped line segments; and a step of detecting one or more road lines in the second disparity map of the plural road line regions of interest.

摘要翻译： 本发明公开了一个道路线检测方法及道路线检测装置。 道路线检测方法包括：获得第一视差图包括一个或多个道路区域和对应的V-视差图像的步骤; 顺序地检测多个斜线段对应的V-视差图像中的雅丁到差异的大到小命令和一个大到小顺序V值的，以用作多个顺序相邻的道路表面上的步骤; 获取的对应于多个倾斜线段兴趣多个道路的行区域的第二视差图的步骤; 和在感兴趣的多个道路的行区域的第二视差图检测一个或多个道路线路的步骤。

公开(公告)号：EP2720193A3

公开(公告)日：2017-12-13

申请号：EP13181638.1

申请日：2013-08-26

申请人： Ricoh Company, Ltd.

发明人： You, Ganmei ,  Lu, Yaojie ,  Shi, Zhongchao ,  Wang, Gang

IPC分类号： G06T7/00 ,  G08G1/00 ,  G06T7/20

CPC分类号： G06T7/70 ,  G06T2207/10021 ,  G06T2207/30252 ,  G06T2207/30256

摘要： Disclosed are a method and a system for detecting an uneven road surface. The method comprises a step of adopting two depth cameras to capture a depth map, and calculating a disparity map; a step of obtaining a V-disparity map by adopting a V-disparity algorithm based on the disparity map; a step of finding an oblique line corresponding to the road surface so as to let a direction of the oblique line be a direction of the road surface; a step of calculating a projection of cross sectional surface of the road surface based on the disparity map and coordinates of disappearance points; and a step of fitting the projection of the cross sectional surface of the road surface to obtain a road line on the cross sectional surface so as to extract the road surface, and letting points located below the fitted road line be road points.

摘要翻译： 公开了一种用于检测不平坦路面的方法和系统。 该方法包括采用两台深度相机拍摄深度图并计算视差图的步骤; 通过基于视差图采用V视差算法来获得V视差图的步骤; 找到与路面对应的斜线以使斜线的方向成为路面的方向的步骤; 基于视差图和消失点的坐标来计算路面的横截面的投影的步骤; 以及拟合路面横截面的投影以获得横截面上的道路线以提取路面，并且使位于拟合道路线下方的点成为道路点的步骤。

公开(公告)号：EP2728546A3

公开(公告)日：2015-04-08

申请号：EP12192685.1

申请日：2012-11-14

申请人： Ricoh Company, Ltd.

发明人： You, Ganmei ,  Lu, Yaojie ,  Zheng, Jichuan

IPC分类号： G06T7/00 ,  G06K9/00

摘要： Disclosed are a method and a system for detecting an object on a road. The method comprises a step of simultaneously capturing two depth maps, and then calculating a disparity map; a step of obtaining, based on the disparity map, a V-disparity image by adopting a V-disparity algorithm; a step of detecting an oblique line in the V-disparity image, and then removing points in the disparity map, corresponding to the oblique line so as to acquire a sub-disparity map excluding the road; a step of detecting plural vertical lines in the V-disparity image, and then extracting, for each of the plural vertical lines, points corresponding to this vertical line from the sub-disparity map as an object sub-disparity map corresponding to this vertical line; and a step of merging any two rectangular areas of the object sub-disparity maps approaching each other, into a rectangular object area.

摘要翻译： 公开了一种用于在道路上检测物体的方法和系统。 该方法包括同时捕获两个深度图，然后计算视差图的步骤; 基于视差图，通过采用V视差算法获得V视差图像的步骤; 检测V视差图像中的斜线，然后去除视差图中与斜线对应的点，以便获取不包括道路的子视差图; 检测V-视差图像中的多个垂直线的步骤，然后针对多个垂直线中的每一个提取与该副视差图对应的该垂直线的点作为与该垂直线对应的对象子视差图 ; 以及将彼此靠近的对象子视差映射的任何两个矩形区域合并为矩形对象区域的步骤。

公开(公告)号：EP2779025A3

公开(公告)日：2015-01-07

申请号：EP14158802.0

申请日：2014-03-11

申请人： Ricoh Company, Ltd.

发明人： You, Ganmei ,  Chen, Chao ,  Shi, Zhongchao ,  Lu, Yaojie ,  Wang, Gang

IPC分类号： G06K9/00

CPC分类号： G06K9/00798

摘要： A method and an apparatus for detecting a road edge are disclosed. The method includes obtaining a disparity map with a road region and a corresponding V-disparity map; extracting, from the V-disparity map, a road line Lv; obtaining, based on the road line Lv extracted from the V-disparity map, a sub-U-disparity map with the road; and self-adaptively extracting, from the sub-U-disparity map, the road edge. According to the method and the apparatus of an embodiment of the present invention, it is possible to detect the road edge accurately, even when there is not accurate road surface information.

公开(公告)号：EP2728546A2

公开(公告)日：2014-05-07

申请号：EP12192685.1

申请日：2012-11-14

申请人： Ricoh Company, Ltd.

发明人： You, Ganmei ,  Lu, Yaojie ,  Zheng, Jichuan

IPC分类号： G06T7/00 ,  G06K9/00

摘要： Disclosed are a method and a system for detecting an object on a road. The method comprises a step of simultaneously capturing two depth maps, and then calculating a disparity map; a step of obtaining, based on the disparity map, a V-disparity image by adopting a V-disparity algorithm; a step of detecting an oblique line in the V-disparity image, and then removing points in the disparity map, corresponding to the oblique line so as to acquire a sub-disparity map excluding the road; a step of detecting plural vertical lines in the V-disparity image, and then extracting, for each of the plural vertical lines, points corresponding to this vertical line from the sub-disparity map as an object sub-disparity map corresponding to this vertical line; and a step of merging any two rectangular areas of the object sub-disparity maps approaching each other, into a rectangular object area.

公开(公告)号：EP2677462A1

公开(公告)日：2013-12-25

申请号：EP13165616.7

申请日：2013-04-26

申请人： Ricoh Company, Ltd.

发明人： Lu, Yaojie ,  You, Ganmei ,  Shi, Zhongchao ,  Wang, Gang

IPC分类号： G06K9/00 ,  G06T7/00

CPC分类号： G06K9/00805 ,  G06K9/00369 ,  G06T7/11 ,  G06T7/174 ,  G06T2207/10012 ,  G06T2207/10021 ,  G06T2207/10028

摘要： A method and an apparatus are used for segmenting an object area, the method comprising: a step of obtaining a disparity map of a scene with an object; a step of obtaining one or more candidate object areas with an object outline from the disparity map; a step of obtaining a disparity range of the object within each of the candidate object areas; a step of obtaining object segmentation areas for segmenting the objects within each of the candidate object areas, based on points within the candidate object area that have a disparity value less than a minimum value of the disparity range of the object within the candidate object area; and a step of segmenting the candidate object areas by the object segmentation areas and obtaining a segmentation result of the object area.

摘要翻译： 一种方法和装置用于分割对象区域，该方法包括：获取场景与对象的视差图的步骤; 从视差图获取具有对象轮廓的一个或多个候选对象区域的步骤; 获取每个候选对象区域内的对象的视差范围的步骤; 基于所述候选对象区域内具有视差值小于候选物体区域内物体的视差范围的最小值的点，获得用于分割每个候选对象区域内的对象的对象分割区域的步骤; 以及通过对象分割区域分割候选对象区域并获得对象区域的分割结果的步骤。