Vehicle target sensing method and system based on looking-around fisheye camera
A perception system and camera technology, which is applied in the field of vehicle target perception methods and systems, can solve the problems of inability to accurately perceive the position and posture of vehicle targets, collision risks, etc., and achieve the effects of reducing the risk of misperception, improving safety, and improving reliability
Pending Publication Date: 2022-08-02
CHONGQING CHANGAN AUTOMOBILE CO LTD
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the position coordinates of the vehicle target given by this method can only give the relative position coordinates of a vehicle target as a whole, and cannot accurately perceive the position and posture of the vehicle target relative to the vehicle. If the vehicle is controlled based on this information, it is easy to cause the risk of collision
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0029] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0030] A vehicle target perception method based on a surround-view fisheye camera, the flowchart of which is as follows figure 1 shown, including the following steps:
[0031] S1: Surround-looking fisheye camera collects the original image of the vehicle target in real time;
[0032] S2: Obtain the original image of the vehicle target, and preprocess the original image;
[0033] S3: Input the preprocessed image into the trained vehicle target detection neural network model and wheel target detection network model, and extract the vehicle target bounding box and wheel bounding box information in the image;
[0034] S4: Calculate the position coordinates of the vehicle target and the position coordinates of the front and rear wheels of the vehicle target according to the obtained vehicle target bounding box and wheel bounding box inform...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a vehicle target sensing method and system based on an all-round fisheye camera. The method specifically comprises the following steps that the all-round fisheye camera collects an original image of a vehicle target in real time; obtaining an original image of the vehicle target, and preprocessing the original image; inputting the preprocessed image into a trained vehicle target detection neural network model and a trained wheel target detection network model, and extracting vehicle target bounding box information and wheel bounding box information in the image; calculating position coordinates of the vehicle target and position coordinates of front and rear wheels of the vehicle target according to the obtained information of the vehicle target bounding box and the wheel bounding box; and calculating the course angle of the vehicle target according to the position coordinates of the front and rear wheels so as to complete the perception of the vehicle target. The method can accurately sense the position posture of the vehicle target relative to the vehicle, and improves the safety of automatic driving.
Description
technical field [0001] The invention belongs to the technical field of automatic driving, and particularly relates to a vehicle target perception method and system based on a surround-view fisheye camera. Background technique [0002] With the development of automobile intelligence, automobile intelligent products are developing from traditional advanced driver assistance systems to automatic driving systems, and application scenarios are gradually upgraded from simple to complex. For example, lane departure warning system, automatic emergency braking system, cruise control system, adaptive cruise system, limited automatic driving system, etc. These systems mainly use sensors such as millimeter-wave radar, lidar, and cameras to capture driving environment information, and then use perception algorithms to obtain information such as lane lines, vehicle targets, and pedestrian targets. [0003] The existing general perception method for vehicle targets is to regard the vehicl...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06V20/56G06V10/82G06T5/50
CPCG06V20/56G06V10/82G06T5/50G06V2201/07G06T2207/20221
Inventor 李涛任凡谭伟高翔王宽
Owner CHONGQING CHANGAN AUTOMOBILE CO LTD