公开(公告)号：US10276040B2

公开(公告)日：2019-04-30

申请号：US15322225

申请日：2015-04-29

Applicant: Connaught Electronics Ltd.

Inventor： Damien Dooley ,  Martin Glavin ,  Edward Jones ,  Liam Kilmartin

IPC: B60Q9/00 ,  G01S5/16 ,  G06K9/00 ,  G06K9/46 ,  G08G1/04 ,  G08G1/16 ,  B60K31/00 ,  G01S11/12 ,  G01S3/786 ,  G06T7/246

Abstract: The invention relates to a method for tracking a target vehicle (9) approaching a motor vehicle (1) by means of a camera system (2) of the motor vehicle (1). A temporal sequence of images (10) of an environmental region of the motor vehicle (1) is provided by means of at least one camera (3) of the camera system (2). The target vehicle (9) is detected in an image (10) of the sequence by means of an image processing device (5) of the camera system (5) based on a feature of a front (11) or of a rear of the target vehicle (9) and then the target vehicle (9) is tracked over subsequent images (10) of the sequence based on the detected feature. Wherein at least a predetermined feature of a lateral flank (14) of the target vehicle (9) is detected in one of the subsequent images (10) of the sequence by the image processing device (5), and after detection of the feature of the lateral flank (14), the target vehicle (9) is tracked over further images (10) of the sequence based on the feature of the lateral flank (14).

公开(公告)号：US09340156B2

公开(公告)日：2016-05-17

申请号：US14686160

申请日：2015-04-14

Applicant: Connaught Electronics Ltd.

Inventor： Diarmad O'Cualain ,  Martin Glavin ,  Edward Jones

IPC: H04N7/18 ,  B60R1/00 ,  G06K9/00 ,  G06K9/20 ,  G06K9/46

CPC classification number: B60R1/00 ,  B60R2300/10 ,  G06K9/00805 ,  G06K9/2036 ,  G06K9/4604 ,  H04N7/183

Abstract: A method for detecting an object in an environmental region of a motor vehicle by a camera system of the motor vehicle is disclosed. A first image of the environmental region is provided by a camera of the camera system with illumination of the environmental region using a light source of the motor vehicle such that a shadow of the object caused by the illumination is depicted in the first image. A second image of the environmental region is also provided by the camera in an inactive state of the light source, in which the illumination is omitted. Then, a subtraction image is generated by subtracting one of the images from the other image by an image processing device of the camera system. Then, an edge image is generated from the subtraction image, and the object is detected based on edges of the edge image by the image processing device.

Abstract translation: 公开了一种通过机动车辆的照相机系统检测机动车辆的环境区域中的物体的方法。 环境区域的第一图像由照相机系统的照相机利用机动车辆的光源照射环境区域，以便在第一图像中描绘由照明产生的物体的阴影。 环境区域的第二图像也由照相机在光源的无效状态下提供，其中省略了照明。 然后，通过相机系统的图像处理装置从另一图像中减去一个图像来生成减法图像。 然后，从减法图像生成边缘图像，并且基于图像处理装置的边缘图像的边缘检测对象。

公开(公告)号：US20170162042A1

公开(公告)日：2017-06-08

申请号：US15322225

申请日：2015-04-29

Applicant: Connaught Electronics Ltd.

Inventor： Damien Dooley ,  Martin Glavin ,  Edward Jones ,  Liam Kilmartin

IPC: G08G1/04 ,  G01S5/16 ,  G06T7/246 ,  B60Q9/00 ,  G06K9/00 ,  G06K9/46 ,  G01S11/12 ,  G08G1/16

CPC classification number: G08G1/04 ,  B60K31/0008 ,  B60K2031/0016 ,  B60Q9/008 ,  B60R2300/108 ,  B60R2300/307 ,  B60R2300/8066 ,  B60R2300/8093 ,  G01S3/786 ,  G01S5/16 ,  G01S11/12 ,  G06K9/00798 ,  G06K9/00805 ,  G06K9/4633 ,  G06T7/248 ,  G06T2207/30261 ,  G08G1/16 ,  G08G1/166 ,  G08G1/167

Abstract: The invention relates to a method for tracking a target vehicle (9) approaching a motor vehicle (1) by means of a camera system (2) of the motor vehicle (1). A temporal sequence of images (10) of an environmental region of the motor vehicle (1) is provided by means of at least one camera (3) of the camera system (2). The target vehicle (9) is detected in an image (10) of the sequence by means of an image processing device (5) of the camera system (5) based on a feature of a front (11) or of a rear of the target vehicle (9) and then the target vehicle (9) is tracked over subsequent images (10) of the sequence based on the detected feature. Wherein at least a predetermined feature of a lateral flank (14) of the target vehicle (9) is detected in one of the subsequent images (10) of the sequence by the image processing device (5), and after detection of the feature of the lateral flank (14), the target vehicle (9) is tracked over further images (10) of the sequence based on the feature of the lateral flank (14).