公开(公告)号：US06369701B1

公开(公告)日：2002-04-09

申请号：US09892740

申请日：2001-06-28

申请人： Takashi Yoshida ,  Nobuhiko Yasui ,  Atsushi Iisaka ,  Akira Ishida

发明人： Takashi Yoshida ,  Nobuhiko Yasui ,  Atsushi Iisaka ,  Akira Ishida

IPC分类号： B60Q100

CPC分类号： B62D15/0275 ,  B60R1/00 ,  B60R2300/105 ,  B60R2300/107 ,  B60R2300/305 ,  B60R2300/60 ,  B60R2300/802 ,  B60R2300/806 ,  B60R2300/8086 ,  G06T5/006 ,  G06T15/20

摘要： In a rendering device Urnd, a CPU generates a lower viewpoint image based on captured images captured by image capture devices both securely mounted to a vehicle. Here, the lower viewpoint image shows a region rear of the vehicle viewed from those image capture devices. Then, the CPU executes viewpoint conversion processing onto the captured images, and generates an upper viewpoint image. Here, the upper viewpoint image shows also the region rear of the vehicle but viewed from a camera virtually placed above the vehicle. The CPU then stitches the upper and lower viewpoint images together to generate a drive assistant image. In such manner, the drive assistant image generated by the rendering device Urnd becomes informative enough for a driver of the vehicle.

摘要翻译： 在渲染设备Urnd中，CPU基于被牢固地安装到车辆的图像捕获设备捕获的捕获图像来生成较低的视点图像。 这里，下视点图像示出从那些图像捕获装置观看的车辆后方的区域。 然后，CPU对拍摄图像执行视点转换处理，并生成上视点图像。 这里，上视点图像还示出了车辆的区域后方，但是从实际上放置在车辆上方的相机观察。 然后，CPU将上下视点图像缝合在一起以产生驱动辅助图像。 以这种方式，由渲染设备Urnd生成的驱动辅助图像对于车辆的驾驶者来说变得足够信息。

公开(公告)号：US06539288B2

公开(公告)日：2003-03-25

申请号：US09862312

申请日：2001-05-23

申请人： Akira Ishida ,  Atsushi Iisaka ,  Takashi Yoshida

发明人： Akira Ishida ,  Atsushi Iisaka ,  Takashi Yoshida

IPC分类号： G05D100

CPC分类号： H04N7/18 ,  B60R1/00 ,  B60R2300/105 ,  B60R2300/30 ,  B60R2300/302 ,  B60R2300/304 ,  B60R2300/404 ,  B60R2300/70 ,  B60R2300/802 ,  H04N7/181

摘要： In a rendering device, a processor receives surrounding images from image capture devices placed in a vehicle, and based thereon, generates a vehicle overlaid image. The processor also receives a rudder angle &thgr; from a rudder angle sensor placed in the vehicle. Then, based on the received rudder angle &thgr;, the processor calculates a 3D predictive trajectory of the vehicle, and on the vehicle overlaid image, renders the calculated 3D predictive trajectories thereby generating a display image. Therefore, the display image generated by the rendering device is driver-friendly.

摘要翻译： 在再现装置中，处理器从放置在车辆中的图像捕获装置接收周围图像，并且基于此，生成车辆重叠图像。 处理器还从放置在车辆中的舵角传感器接收舵角θ。 然后，基于接收到的方向舵角θ，处理器计算车辆的3D预测轨迹，并且在车辆重叠图像上，呈现所计算的3D预测轨迹从而生成显示图像。 因此，由再现装置生成的显示图像是驾驶者友好的。

公开(公告)号：US07304651B2

公开(公告)日：2007-12-04

申请号：US09861548

申请日：2001-05-22

申请人： Akira Ishida ,  Atsushi Iisaka ,  Takashi Yoshida

发明人： Akira Ishida ,  Atsushi Iisaka ,  Takashi Yoshida

IPC分类号： G09G5/00 ,  B60Q1/00 ,  G09F9/00 ,  G08G1/00 ,  G08G1/017 ,  G08G1/054 ,  H04N7/18 ,  H04N9/47

CPC分类号： H04N7/18 ,  B60R1/00 ,  B60R2300/105 ,  B60R2300/30 ,  B60R2300/302 ,  B60R2300/304 ,  B60R2300/404 ,  B60R2300/70 ,  B60R2300/802 ,  H04N7/181

摘要： In a rendering device Urnd1, a processor 1 receives a front-left image Sfl and a front-right image Sfr at regular intervals, and therefrom, cuts out a partial front-left image and a partial front-right image for overlay onto a mask image. The processor 1 then overlays, with a predetermined timing, any one of vehicle model images varying in size onto the mask image. Here, the vehicle model image is overlaid in such manner as to create a moving picture as if the vehicle model image is moving forward. With such vehicle model image, a driver of a vehicle can instantaneously recognize that a drive assistant device is now showing the front side of the vehicle.

摘要翻译： 在渲染设备Urnd 1中，处理器1以规则的间隔接收前左图像Sfl和前右图像Sfr，并从中切出部分前左图像和部分右前图像以覆盖到 掩模图像。 然后，处理器1以预定的定时将大小变化的车辆模型图像中的任何一个叠加到掩模图像上。 这里，车辆模型图像以如同车辆模型图像向前移动的方式创建运动图像的方式被覆盖。 利用这种车辆模型图像，车辆的驾驶员可以立即识别出驾驶辅助装置正在显示车辆的前侧。

公开(公告)号：US5904725A

公开(公告)日：1999-05-18

申请号：US637417

申请日：1996-04-25

申请人： Atsushi Iisaka ,  Nobuhiko Yasui ,  Mamoru Kaneko

发明人： Atsushi Iisaka ,  Nobuhiko Yasui ,  Mamoru Kaneko

IPC分类号： G01S5/16 ,  G06T5/00 ,  G06T7/00 ,  G06F165/00

CPC分类号： G01S5/16 ,  G06K9/00798 ,  G06T7/0044 ,  G06T7/0083 ,  G06T2207/10016 ,  G06T2207/30256

摘要： A local positioning device detects a local position of an automobile advancing along a lane formed on a road in a direction. A digital image data indicative of an advancing direction view observed from the automobile in the advancing direction is applied to the local position device. An edge extractor, a threshold generator, and a contour extractor are operable for extracting a contour data indicative of the lane from the image signal. A coordinate convertor converts the contour data into a plane view data indicating the dimensions of extracted contours correctly. A matching operator matches a line or an arc with each of extracted contour. A lane mark contour extractor selectively extracts a pair of extracted contours corresponding to the lane on which the automobile is located.

摘要翻译： 本地定位装置检测沿着在道路上形成的车道沿方向前进的汽车的局部位置。 将表示从行驶方向观察到的前进方向图的数字图像数据应用于本地位置装置。 边缘提取器，阈值生成器和轮廓提取器可用于从图像信号中提取表示车道的轮廓数据。 坐标转换器将轮廓数据转换为指示提取轮廓的尺寸正确的平面视图数据。 匹配的操作符将线或弧与每个提取的轮廓相匹配。 车道标记轮廓提取器选择性地提取对应于汽车所在的车道的一对提取轮廓。

公开(公告)号：US06483429B1

公开(公告)日：2002-11-19

申请号：US09692540

申请日：2000-10-20

申请人： Nobuhiko Yasui ,  Atsushi Iisaka

发明人： Nobuhiko Yasui ,  Atsushi Iisaka

IPC分类号： B60Q100

CPC分类号： B60R1/00 ,  B60Q9/005 ,  B60R2300/30 ,  B60R2300/302 ,  B60R2300/305 ,  B60R2300/307 ,  B60R2300/404 ,  B60R2300/607 ,  B60R2300/806 ,  B60R2300/8093 ,  B60T2201/10 ,  G01S5/16 ,  G06K9/00812

摘要： A parking assistance system is mounted in a vehicle, and has an image pick-up with a single camera, an image processing device, a display, a steering angle sensor, a wheel speed sensor, and a pulse counter. An A/D converter subjects two analog images picked up by the image pick-up at different locations to A/D conversion, and sends the results to frame memories. A CPU uses the digital image data and the transition data of the vehicle to calculate object data and a distance from the vehicle to the 3D object. An image thus obtained is then converted into a view from the above. The view from the above is stored in a frame memory, and outputted to the display under the control of a controller for display.

摘要翻译： 车辆中安装有停车辅助系统，并且具有单个照相机，图像处理装置，显示器，转向角传感器，车轮速度传感器和脉冲计数器的图像拾取器。 A / D转换器将由不同位置的图像拾取器拾取的两个模拟图像进行A / D转换，并将结果发送到帧存储器。 CPU使用数字图像数据和车辆的转换数据来计算对象数据和从车辆到3D对象的距离。 然后将如此获得的图像从上述转换成视图。 从上述的视图被存储在帧存储器中，并且在控制器的控制下输出到显示器以进行显示。

公开(公告)号：US06546118B1

公开(公告)日：2003-04-08

申请号：US09228401

申请日：1999-01-12

申请人： Atsushi Iisaka ,  Nobuhiko Yasui

发明人： Atsushi Iisaka ,  Nobuhiko Yasui

IPC分类号： G06K900

CPC分类号： G06K9/00798 ,  G06T7/246 ,  G06T7/73

摘要： An image processing device for a vehicle according to the present invention has an edge extractor for emphasizing edges of a traffic lane in a front road image signal generated by a digital image pick-up device, a threshold setter for setting a threshold to extract the edges of the traffic lane based on the result detected by the edge extractor, an outline extractor for extracting an outline of the traffic lane to generate an outline data based on the result detected by the edge extractor and the threshold set by the threshold setter, a noise eliminator for eliminating noise components from the outline data, a white line detector for detecting lane-indicating lines based on the outline data from which the noise components have been eliminated, an attitude change detector for estimating the attitude of the vehicle based on the detected lane-indicating lines, and a target advance direction detector for detecting a target direction to advance the vehicle based on the result obtained by the white line detector and the estimated attitude of the vehicle. According to the image processing apparatus, the target direction to advance the vehicle can be exactly measured, even if the helix angle (i.e. pitch angle) changes.

摘要翻译： 根据本发明的用于车辆的图像处理装置具有用于强调由数字图像拾取装置产生的前道路图像信号中的行车道边缘的边缘提取器，用于设置阈值以提取边缘的阈值设置器 基于由边缘提取器检测到的结果的轮廓提取器，用于基于由边缘提取器检测到的结果和由阈值设置器设置的阈值来提取行车道的轮廓以生成轮廓数据的轮廓提取器，噪声 用于从轮廓数据中消除噪声分量的消除器，用于基于已经消除了噪声分量的轮廓数据来检测车道指示线的白线检测器，用于基于检测到的车道估计车辆的姿态的姿态变化检测器 - 指示线，以及目标提前方向检测器，用于基于由th获得的结果来检测目标方向以推进车辆 e白线检测器和车辆的估计姿态。 根据图像处理装置，即使螺旋角（即俯仰角）发生变化，也可以精确地测量推进车辆的目标方向。

公开(公告)号：US6091833A

公开(公告)日：2000-07-18

申请号：US917815

申请日：1997-08-27

申请人： Nobuhiko Yasui ,  Atsushi Iisaka ,  Mamoru Kaneko

发明人： Nobuhiko Yasui ,  Atsushi Iisaka ,  Mamoru Kaneko

IPC分类号： G06T1/00 ,  G05D1/02 ,  G06K9/00 ,  G06T5/00 ,  G06T7/00 ,  G08G1/04 ,  G08G1/16

CPC分类号： G06K9/00798 ,  G05D1/0246 ,  G06T7/0042 ,  G06T7/0044 ,  G06T7/0083 ,  G08G1/04 ,  G08G1/161 ,  B60T2201/08 ,  B60T2201/089 ,  G06T2207/10016 ,  G06T2207/30256

摘要： An apparatus and method for correctly determining the position of a vehicle in a traffic lane by obtaining correct information about the position of the traffic lane without being affected by variations in the road surface, weather, time of day, or such imaging conditions as fixed or moving lighting, are provided. An edge signal of a high spatial frequency component and a luminance signal of a low spatial frequency component of a digital image signal representing the view of the local area to the front of a vehicle are extracted. A road contour signal is then extracted from the edge signal, and a road region signal is extracted from the luminance signal. The position of the lane Sre is then detected with high precision by evaluating the lane contour Sre based on the road region signal Srr and lane contour data Sre.

摘要翻译： 通过获取关于行车道的位置的正确信息而不受路面，天气，时间的变化或诸如固定或 提供移动照明。 提取高空间频率分量的边缘信号和表示车辆前方的局部区域的视图的数字图像信号的低空间频率分量的亮度信号。 然后从边缘信号中提取道路轮廓信号，并从亮度信号中提取道路区域信号。 然后通过基于道路区域信号Srr和车道轮廓数据Sre评估车道轮廓Sre，高精度地检测车道Sre的位置。

公开(公告)号：US06731332B1

公开(公告)日：2004-05-04

申请号：US09238611

申请日：1999-01-28

申请人： Nobuhiko Yasui ,  Atsushi Iisaka ,  Noboru Nomura

发明人： Nobuhiko Yasui ,  Atsushi Iisaka ,  Noboru Nomura

IPC分类号： H04N718

CPC分类号： H04N7/183 ,  G06K9/00798 ,  G06T7/246

摘要： In an image processing apparatus carried on a vehicle, an image of a scene around the vehicle, which is photographed by an image pick-up, is inputted into a white line failure extractor. Then, outline points of a white line on a road and a guardrail, which is parallel to the white line, are sought. Further, the outline points are detected as a polynomial. Moreover, the image photographed by the image pick-up is also inputted into an intermittent image obtainer. Then, the images from the image pick-up are intermittently obtained at constant periodical intervals, and further the images are compressed using JPEG compression method. Further, in a synthesizer, the feature data from the white line feature extractor and the image data from the intermittent image obtainer are aligned in order of time. Moreover, the image data and feature data, which are aligned in order of time, are conserved within a given constant period by a recorder. Thus, accurate data as to condition of the road can be recorded using a small capacity memory.

摘要翻译： 在搭载在车辆上的图像处理装置中，由摄像机摄影的车辆周围的场景的图像被输入白线故障提取器。 然后，寻求与白线平行的道路上的白线的轮廓点和护栏。 此外，轮廓点被检测为多项式。 此外，由图像拾取器拍摄的图像也被输入到间歇图像获取器中。 然后，以恒定的周期间隔间歇地获得来自图像拾取器的图像，并且使用JPEG压缩方法进一步压缩图像。 此外，在合成器中，来自白线特征提取器的特征数据和来自间歇图像获取器的图像数据按时间顺序排列。 此外，由时间顺序排列的图像数据和特征数据由记录器在给定的恒定周期内保存。 因此，可以使用小容量存储器记录关于道路状况的准确数据。

公开(公告)号：US06493458B2

公开(公告)日：2002-12-10

申请号：US09216957

申请日：1998-12-21

申请人： Nobuhiko Yasui ,  Atsushi Iisaka ,  Mamoru Kaneko

发明人： Nobuhiko Yasui ,  Atsushi Iisaka ,  Mamoru Kaneko

IPC分类号： G06K900

CPC分类号： G06K9/00798 ,  B60T2201/08 ,  B60T2201/089 ,  G05D1/0246 ,  G06T7/12 ,  G06T7/73 ,  G06T7/74 ,  G06T2207/10016 ,  G06T2207/30256 ,  G08G1/04 ,  G08G1/161

摘要： An apparatus and method for correctly determining the position of a vehicle in a traffic lane by obtaining correct information about the position of the traffic lane without being affected by variations in the road surface, weather, time of day, or such imaging conditions as fixed or moving lighting, are provided. An edge signal of a high spatial frequency component and a luminance signal of a low spatial frequency component of a digital image signal representing the view of the local area to the front of a vehicle are extracted. A road contour signal is then extracted from the edge signal, and a road region signal is extracted from the luminance signal. The position of the lane Sre is then detected with high precision by evaluating the lane contour Sre based on the road region signal Srr and lane contour data Sre.

摘要翻译： 通过获取关于行车道的位置的正确信息而不受路面，天气，时间的变化或诸如固定或 提供移动照明。 提取高空间频率分量的边缘信号和表示车辆前方的局部区域的视图的数字图像信号的低空间频率分量的亮度信号。 然后从边缘信号中提取道路轮廓信号，并从亮度信号中提取道路区域信号。 然后通过基于道路区域信号Srr和车道轮廓数据Sre评估车道轮廓Sre，高精度地检测车道Sre的位置。

公开(公告)号：US06344805B1

公开(公告)日：2002-02-05

申请号：US09560560

申请日：2000-04-28

申请人： Nobuhiko Yasui ,  Atsushi Iisaka ,  Yoshiki Ueyama

发明人： Nobuhiko Yasui ,  Atsushi Iisaka ,  Yoshiki Ueyama

IPC分类号： B60Q148

CPC分类号： B62D15/0275 ,  B60T2201/10 ,  B62D15/027 ,  G08G1/168

摘要： A vehicle position calculation unit (1) calculates a position of a vehicle on a plan view, and an estimate locus calculation unit (2) calculates an estimate movement locus of a vehicle body when the vehicle is moved at a maximum steering angle with respect to the vehicle position, and a display unit (3) displays the vehicle position calculated by the vehicle position calculation unit (1) together with the estimate locus of the vehicle body calculated by the estimate locus calculation unit (2). Thus, even a beginner driver of the vehicle can easily and quickly park the vehicle with accuracy and safety while monitoring the estimate movement locus of the vehicle on the display screen.

摘要翻译： 车辆位置计算单元在平面图中计算车辆的位置，并且估计轨迹计算单元计算当车辆相对于车辆位置以最大转向角移动时的车身的估计移动轨迹，并且 显示单元显示由车辆位置计算单元计算的车辆位置与由估计轨迹计算单元计算出的车身的估计轨迹。 因此，即使车辆的初级驾驶员也可以在监视车辆在显示画面上的估计移动轨迹的同时，以准确和安全的方式容易且快速地停放车辆。