Lane line recognition method based on seamless stitching of multi-source reverse perspective images
A technology of lane line recognition and seamless splicing, applied in character and pattern recognition, instruments, computer parts, etc., can solve the problems of limited and unsolvable recognition field of view, and achieve the effect of overcoming the limited field of view
Inactive Publication Date: 2018-02-02
BEIJING UNION UNIVERSITY
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The purpose of the present invention is to overcome the shortcomings of the existing lane line recognition methods, such as the limited recognition field of view and the inability to solve the problem of lane line recognition on multi-lane and large-curvature curves, and proposes a lane line based on seamless splicing of multi-source reverse perspective images recognition methods
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 approach
[0032] 1. According to the requirements of step 1, install a monocular camera at the left and right reversing mirrors of the vehicle and under the interior rearview mirror. The platform used in this example has already installed the required devices, and the experiment can be carried out only with a slight adjustment.
[0033] 2. Follow the detailed steps of steps 2, 3, and 4 to implement. The parameters involved are: the width of the test lane is 300cm; the width of the lane line is 15cm, and the width of the vehicle is 170cm; The horizontal axis width of the field of view of the picture is 450cm, and the vertical axis width is 2000cm; the horizontal axis width of the bird's eye view is 170cm, and the vertical axis width is 1500cm.
[0034] 3. The correct rate of lane line recognition is 100%.
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
A lane line recognition method based on seamless stitching of multi-source reverse perspective images belongs to the field of computer vision and the field of safe intelligent transportation. Firstly, the video images are collected by the cameras installed at the interior rearview mirror of the vehicle and the cameras at the left and right side mirrors of the vehicle, and the original images of these three cameras are preprocessed respectively, and then the reverse perspective transformation is performed, and then the three reverse perspective images are Perform seamless splicing to obtain an image, and finally perform lane line recognition on the image. The lane line recognition method based on the seamless splicing of multi-source reverse perspective images can solve the problems of multi-lane line recognition and large curvature curve recognition, and is suitable for smart car visual navigation and lane departure warning.
Description
Technical field: [0001] The invention discloses a lane line recognition method based on seamless splicing of multi-source reverse perspective images, which belongs to the field of computer vision and the field of safe intelligent transportation. Background technique: [0002] With the continuous development of social economy and science and technology, automobiles have become a part of people's lives as a means of transportation, and safe driving and intelligent driving have become the direction people are pursuing. Whether it is a lane departure warning system or an unmanned smart car visual navigation, lane line recognition is required, but multi-lane recognition and large-curvature curve recognition are still difficult points in lane line recognition. How to solve multi-lane and large-curvature curve recognition become more and more important. [0003] The Chinese invention patent with the publication number CN102806913A discloses a new type of lane deviation detection m...
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 Patents(China)
IPC IPC(8): G06K9/60
Inventor 袁家政刘宏哲鲍泓郑永荣
Owner BEIJING UNION UNIVERSITY