Image-based traffic state detection method
A technology of traffic status and detection method, applied in the field of traffic detection, can solve problems such as low data utilization rate
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0094] An image-based traffic state detection method, comprising:
[0095] Obtain traffic monitoring video, divide it according to the resolution, and determine the video clips that reach the preset resolution;
[0096] Determining the section area to be detected through grayscale impact detection and image binarization of the video clip;
[0097] Segmenting the video clips through multi-scale images to determine vehicle features, and constructing a three-dimensional model of the vehicle to determine the vehicle bottom area;
[0098] The perspective projection of the monitoring video is converted to an orthographic projection through projection transformation (generally, the surveillance video is located in front of the road side, and the shooting image has geometric deformation. Projection transformation is to transform the image into a vertical upper shooting effect, eliminating the estimation error caused by the deformation of the road and vehicles ), and determine the tra...
Embodiment 2
[0113] Preferably, the method also includes:
[0114] Docking with the data source of traffic surveillance video; among them,
[0115] The data sources include: road monitoring system, private road test monitoring equipment and traffic monitoring satellite;
[0116] Obtain traffic monitoring videos from different channels through the data source; wherein,
[0117] The traffic monitoring video includes: tunnel video, road video, bridge video;
[0118] Carry out traffic monitoring videos from different channels.
[0119] The principle of the above-mentioned technical scheme is: the data collection of the present invention utilizes traffic monitoring video and images as data sources, as attached figure 2 shown, but with figure 2 The collected images are original traffic images with relatively low resolution. The distance in the image cannot be detected, so it is necessary to intercept the image to determine the area of the detected road section. The intercepted image is ...
Embodiment 3
[0122] Preferably, the method also includes:
[0123] According to the traffic monitoring video, determine vehicle information and road pavement standards on the road; wherein,
[0124] The vehicle information includes: vehicle quantity, vehicle location, vehicle type and vehicle volume;
[0125] The road pavement standard includes: road width, road lane type and road division standard;
[0126] The road information is determined according to the vehicle information on the road and road surface standards.
[0127] The principle of the above-mentioned technical solution is: the present invention mainly uses the video of traffic monitoring to calculate the degree of traffic congestion, so the most important thing is that the vehicle information and road standards on the road can be clearly known; through the vehicle information and road standards, To calculate the space occupancy rate of the road and the degree of traffic congestion. The definition of road space occupancy is:...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


