公开(公告)号：US20130266226A1

公开(公告)日：2013-10-10

申请号：US13441983

申请日：2012-04-09

申请人： Wende Zhang ,  Jinsong Wang

发明人： Wende Zhang ,  Jinsong Wang

IPC分类号： G06K9/46

CPC分类号： B60W40/04 ,  G06K9/00791

摘要： A method of detecting a clear path of travel. Input images are captured at various time step frames. Clear path probability maps of a current and previous time step frames are generated. A corresponding clear path probability map is generated for the current time step frame derived as a function of the clear path probability map of the previous time step frame and of a corresponding mapping that coherently links the previous time step frame to the current time step frame. A weight-matching map is generated. The probability values of the current time step frame are updated as a function of the corresponding probability map. A current frame probability decision map is generated based on updated probability values of the current time step frame. The clear path in the image of the current time step is identified based on the current frame probability decision map.

摘要翻译： 一种检测行驶路线清晰的方法。 在各种时间步长帧中捕获输入图像。 生成当前和前一时间步长帧的清除路径概率图。 对于作为先前时间步长帧的清除路径概率图的函数导出的当前时间步长帧以及将先前时间步长帧相干地链接到当前时间步长帧的对应映射，生成对应的清除路径概率图。 生成权重匹配图。 根据相应的概率图，更新当前时间步长帧的概率值。 基于当前时间步长帧的更新概率值生成当前帧概率决策图。 基于当前帧概率判定图来识别当前时间步长的图像中的清晰路径。

公开(公告)号：US20130202152A1

公开(公告)日：2013-08-08

申请号：US13366420

申请日：2012-02-06

申请人： Wende Zhang ,  Jinsong Wang

发明人： Wende Zhang ,  Jinsong Wang

IPC分类号： G06K9/60

CPC分类号： G06K9/00791

摘要： A method provides for determining visible regions in a captured image during a nighttime lighting condition. An image is captured from an image capture device mounted to a vehicle. An intensity histogram of the captured image is generated. An intensity threshold is applied to the intensity histogram for identifying visible candidate regions of a path of travel. The intensity threshold is determined from a training technique that utilizes a plurality of training-based captured images of various scenes. An objective function is used to determine objective function values for each correlating intensity value of each training-based captured image. The objective function values and associated intensity values for each of the training-based captured images are processed for identifying a minimum objective function value and associated optimum intensity threshold for identifying the visible candidate regions of the captured image.

摘要翻译： 一种方法用于在夜间照明条件期间确定拍摄图像中的可见区域。 从安装到车辆的图像捕获装置捕获图像。 生成捕获图像的强度直方图。 将强度阈值应用于用于识别行进路径的可见候选区域的强度直方图。 强度阈值由利用多个基于训练的各种场景的拍摄图像的训练技术确定。 目标函数用于确定每个基于训练的捕获图像的每个相关强度值的目标函数值。 处理每个基于训练的拍摄图像的目标函数值和相关联的强度值，以便识别用于识别捕获图像的可见候选区域的最小目标函数值和相关联的最佳强度阈值。

公开(公告)号：US20130136309A1

公开(公告)日：2013-05-30

申请号：US13307236

申请日：2011-11-30

申请人： Wende Zhang ,  Bakhtiar Brian Litkouhi

发明人： Wende Zhang ,  Bakhtiar Brian Litkouhi

IPC分类号： G06K9/00

CPC分类号： G06T7/246 ,  G06T2207/10016 ,  G06T2207/20164 ,  G06T2207/30261

摘要： A method of determining functionality of a vision-based imaging system for a vehicle includes capturing images by the vision-based imaging system. The images include landmarks that are identified in multiple images for distinguishing displacement of landmarks between the images. Edge maps for each of the images are generated. Like regions of at least two edge maps are compared for distinguishing displacement of a landmark between the at least two edge maps. Each pixel location of a first edge map is compared with a same pixel location of a second edge map for determining whether each compared pixel location has a different intensity value. An aggregate number of pixel locations having a different intensity value between the at least two edge maps is determined and compared to a threshold. An error message is generated in response to the aggregate number being less than the threshold.

摘要翻译： 确定用于车辆的基于视觉的成像系统的功能的方法包括通过基于视觉的成像系统捕获图像。 图像包括在多个图像中识别的地标，用于区分图像之间的界标位移。 生成每个图像的边缘图。 与至少两个边缘图相似的区域被比较以区分至少两个边缘图之间的地标的位移。 将第一边缘图的每个像素位置与第二边缘图的相同像素位置进行比较，以确定每个比较的像素位置是否具有不同的强度值。 确定在所述至少两个边缘图之间具有不同强度值的像素位置的总数，并将其与阈值进行比较。 响应于总数小于阈值而生成错误消息。

公开(公告)号：US08447519B2

公开(公告)日：2013-05-21

申请号：US12943247

申请日：2010-11-10

申请人： Chaminda Basnayake ,  Wende Zhang

发明人： Chaminda Basnayake ,  Wende Zhang

IPC分类号： G01C21/10

CPC分类号： G01S19/48 ,  G01C21/30 ,  G01C21/3602 ,  G01S3/7864 ,  G01S19/40

摘要： A method is provided for augmenting GPS data using an in-vehicle vision-based module. A vehicle position is determined utilizing position-related data obtained from a position module. A position error is estimated on a periodic basis. A determination is made whether the position error estimate exceeds a first predetermined error threshold. Tracking data is generated for the vehicle over a course of travel utilizing captured images from the in-vehicle vision based module. The tracking data is integrated with the position-related data to estimate the vehicle position in response to the position error estimate exceeding the first predetermined error threshold. A determination is made whether the position error estimate decreases below a second predetermined error threshold. The vehicle position is re-determined using only the position-related data when the position error estimate decreases below the second predetermined error threshold.

摘要翻译： 提供了一种使用基于车载视觉的模块来增强GPS数据的方法。 使用从位置模块获得的位置相关数据来确定车辆位置。 定期地估计位置误差。 确定位置误差估计是否超过第一预定误差阈值。 利用来自基于车载视觉的模块的捕获图像，在行驶过程中为车辆生成跟踪数据。 跟踪数据与位置相关数据集成，以响应于超过第一预定误差阈值的位置误差估计来估计车辆位置。 确定位置误差估计是否降低到低于第二预定误差阈值。 当位置误差估计值降低到低于第二预定误差阈值时，仅使用位置相关数据重新确定车辆位置。

公开(公告)号：US20120314070A1

公开(公告)日：2012-12-13

申请号：US13157124

申请日：2011-06-09

申请人： Wende Zhang ,  Bakhtiar Brian Litkouhi ,  Jin-Woo Lee

发明人： Wende Zhang ,  Bakhtiar Brian Litkouhi ,  Jin-Woo Lee

IPC分类号： H04N7/18 ,  G06K9/60

CPC分类号： B60W40/00 ,  B60W30/12 ,  B60W30/18163 ,  B60W2420/42 ,  B60W2550/304

摘要： A system and method for accurately estimating a lane in which a vehicle is traveling. A sensor mounted on the vehicle generates sensor data including lane information that is processed by several lane detection sub-systems to generate two or more estimated lanes with corresponding lane confidence information. A combining processor combines the estimated lanes based upon the confidence information to determine a combined estimated lane.

摘要翻译： 一种用于准确估计车辆行驶的车道的系统和方法。 安装在车辆上的传感器产生包括由多个车道检测子系统处理的车道信息的传感器数据，以生成具有对应车道置信度的两个或多个估计车道。 组合处理器基于置信度信息组合估计的车道以确定组合的估计车道。

公开(公告)号：US20120008019A1

公开(公告)日：2012-01-12

申请号：US12830513

申请日：2010-07-06

申请人： Wende Zhang ,  Qi Wu ,  Vijayakumar Bhagavatula

发明人： Wende Zhang ,  Qi Wu ,  Vijayakumar Bhagavatula

IPC分类号： H04N9/64

CPC分类号： G06K9/00791 ,  G06K9/346 ,  G06T5/008 ,  G06T7/90 ,  G06T2207/10024 ,  G06T2207/30252

摘要： A method is provided for removing an illumination generated shadow in a captured image. Each pixel of the captured input image is plotted on a two dimensional logarithmic graph. A linear axis for the plurality of color sets is determined that is substantially orthogonal to a respective illumination direction of each respective color set. A log-chromaticity value of each plotted pixel is projected on the axis. An orientation of the linear axis is selected to minimize an illumination effect and provide optimum separation between each of the respective color sets on the linear axis. Edges in the input image and illumination invariant image domain are identified. The identified edges of the input image are compared to identify edges in the illumination invariant image domain. A determination is made whether a shadow edge is present in response to the comparison. A shadow-reduced image is generated for scene analysis by a vehicle vision-based system.

摘要翻译： 提供了一种用于去除拍摄图像中的照明产生阴影的方法。 捕获的输入图像的每个像素绘制在二维对数图上。 确定多个颜色组的线性轴，其基本上与每个相应颜色组的相应照明方向正交。 每个绘制像素的对数色度值投影在轴上。 选择线性轴的取向以最小化照明效果并且在线性轴上的每个相应颜色集之间提供最佳间隔。 识别输入图像和照明不变图像域中的边缘。 将识别的输入图像的边缘进行比较，以识别照明不变图像域中的边缘。 确定响应于比较是否存在阴影边缘。 通过基于车辆视觉的系统生成用于场景分析的阴影缩小图像。

公开(公告)号：US20110221901A1

公开(公告)日：2011-09-15

申请号：US12721801

申请日：2010-03-11

申请人： Fan Bai ,  Wende Zhang ,  Cem U. Saraydar

发明人： Fan Bai ,  Wende Zhang ,  Cem U. Saraydar

IPC分类号： H04N7/18 ,  G06K9/00

CPC分类号： H04L69/04 ,  H04L67/12 ,  H04L67/18 ,  H04N19/115 ,  H04N19/124 ,  H04N19/162

摘要： A method is provided for video sharing in a vehicle-to-entity communication system. Video data is captured by an image capture device of an event remote from a source entity. A spatial relationship is determined between a location corresponding to the captured event and a location of a remote vehicle. A temporal relationship is determined between a time-stamp of the captured scene data and a current time. A utility value is determined as a function of the spatial relationship and the temporal relationship. A network utilization parameter of a communication network is determined for broadcasting and receiving the scene data. A selected level of compression is applied to the captured scene data as a function of the utility value and available bandwidth. The compressed scene data is transmitted from the source entity to the remote vehicle.

摘要翻译： 提供了一种用于车对实体通信系统中的视频共享的方法。 视频数据由远离源实体的事件的图像捕获设备捕获。 在与捕获的事件相对应的位置和远程车辆的位置之间确定空间关系。 在捕获的场景数据的时间戳与当前时间之间确定时间关系。 确定效用值作为空间关系和时间关系的函数。 确定通信网络的网络利用率参数用于广播和接收场景数据。 作为效用值和可用带宽的函数，所选择的压缩级别被应用于捕获的场景数据。 压缩场景数据从源实体发送到远程车辆。

公开(公告)号：US20100131197A1

公开(公告)日：2010-05-27

申请号：US12276050

申请日：2008-11-21

申请人： Wende Zhang ,  Varsha Sadekar

发明人： Wende Zhang ,  Varsha Sadekar

IPC分类号： G01C21/36 ,  G06K9/00

CPC分类号： G01C21/36 ,  G01C21/3647

摘要： A system and method that provide a video-based vehicle navigation system. The system positions an arrow on the video display that shows the specific turning direction for the vehicle for route guidance purposes. To determine the proper position of the guidance arrow, the process determines a distance from the current vehicle position to the location where the vehicle needs to turn using any suitable information, such as GPS position, range measurements and map information. The process then positions the guidance arrow on the ground at the turning location in world coordinates, and projects the guidance arrow onto the image. The camera can be calibrated to the ground using various techniques, such as an online automatic calibration process that uses detected objects in the scene around the vehicle.

摘要翻译： 一种提供基于视频的车载导航系统的系统和方法。 系统在视频显示器上放置一个箭头，显示车辆的特定转向方向，用于路线引导目的。 为了确定导向箭头的正确位置，该过程使用诸如GPS位置，距离测量和地图信息之类的任何合适的信息确定从当前车辆位置到车辆需要转动的位置的距离。 然后，该过程将引导箭头定位在世界坐标中的转动位置的地面上，并将引导箭头投影到图像上。 摄像机可以使用各种技术进行校准，例如使用车辆周围场景中检测到的物体的在线自动校准过程。

公开(公告)号：US20100021010A1

公开(公告)日：2010-01-28

申请号：US12179645

申请日：2008-07-25

申请人： Wende Zhang ,  Mark J. Wolski

发明人： Wende Zhang ,  Mark J. Wolski

IPC分类号： G06K9/00 ,  H04N7/18

CPC分类号： G06K9/00369 ,  B60K31/0008 ,  G06K9/4647

摘要： A method is provided for detecting a pedestrian exterior to a vehicle by use of a vehicle-mounted image-based monitoring system. An edge-based pedestrian detection analysis is applied to the input image data to determine an edge-based confidence value of a candidate image area being a pedestrian. A motion-based pedestrian detection analysis is applied to the input image data to determine a motion-based confidence value of a candidate image area being a pedestrian. A final confidence value is obtained by combining the edge-based confidence value and motion-based confidence value using weighting factors that are functions of vehicle speed. The final confidence value is compared to a threshold value for determining whether to generate a warning that a pedestrian may be present in the input image data.

摘要翻译： 提供一种通过使用车载图像监视系统来检测车辆的行人外观的方法。 对输入图像数据应用基于边缘的行人检测分析，以确定作为行人的候选图像区域的基于边缘的置信度值。 基于运动的行人检测分析被应用于输入图像数据，以确定作为行人的候选图像区域的基于运动的置信度值。 通过使用作为车辆速度的函数的加权因子组合基于边缘的置信度值和基于运动的置信度值来获得最终置信度值。 将最终置信度值与用于确定是否产生行人可能存在于输入图像数据中的警告的阈值进行比较。

公开(公告)号：US09751529B2

公开(公告)日：2017-09-05

申请号：US13156974

申请日：2011-06-09

申请人： Wende Zhang ,  Bakhtiar Brian Litkouhi ,  Jin-Woo Lee

发明人： Wende Zhang ,  Bakhtiar Brian Litkouhi ,  Jin-Woo Lee

IPC分类号： B62D6/00 ,  B60W30/12 ,  B62D15/02 ,  B60W10/20 ,  B60W40/06

CPC分类号： B60W30/12 ,  B60W10/20 ,  B60W40/06 ,  B60W2420/42 ,  B60W2710/202 ,  B60W2710/207 ,  B60W2720/14

摘要： A method for adjusting a vehicle's position in a roadway lane. A camera mounted on the vehicle generates a current image of the lane and the method identifies a current lane-center line in the current image. A reference image is generated and the method identifies a reference lane-center line in the reference image. The method then calculates an error between the current lane-center line and the reference lane-center line and provide steering commands to adjust the position of the vehicle so that the error is reduced.