Unlicensed Vehicle Violation Detection Method and Detection System
A technology of unlicensed vehicles and detection methods, applied in the field of intelligent traffic management, can solve problems such as road congestion, consumption of large manpower and material resources, etc.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Publication Date
- 2017-01-25
Smart Images

Figure 1 
Figure 2
Abstract
Description
technical field
[0001] The invention belongs to the technical field of intelligent traffic management, and in particular relates to a method for detecting violations of unlicensed vehicles and a detection system thereof. Background technique
[0002] At present, some vehicles on the road do not have license plates, or license plates are lost, or their license plates are defaced or deliberately covered. hold accountable. In order to strengthen the standardized management of vehicles and ensure traffic safety, it is necessary to effectively detect and track unlicensed vehicles on the road, so as to better grasp the driving information on the road.
[0003] For the detection of unlicensed vehicles, manual law enforcement is generally used, and the traffic police set up roadblocks to check the passing vehicles at a certain intersection, but this method consumes a lot of manpower and material resources, and it is easy to cause road congestion. Contents of the invention [000...
Examples
Embodiment Construction
[0034] Such as figure 1 As shown, an unlicensed vehicle violation detection method includes:
[0035] S101. Video image collection: collect video images in real time through cameras along the direction of the lane, where the cameras may be monocular high-definition cameras.
[0036] S102. Unlicensed vehicle detection:
[0037] 1. Detect the vehicle target in the video image, and locate the rectangular body area of the vehicle target:
[0038] The video image is traversed multiple times through the detection window. The size of the detection window is enlarged with the number of traversals. The haar feature value in the detection window is calculated and classified by the structured svm classifier. If the window with the largest classification score exceeds the preset threshold, then This area is the rectangular body area of the vehicle target. The structured svm classifier is composed of vehicle positive sample set with haar feature (rectangular feature) and non-vehicle...