Highway expense clearing method and device

Abstract

The embodiment of the invention provides a highway expense clearing method and device, and the method comprises the steps: obtaining a road condition video photographed by a camera, determining the driving track of a to-be-cleared vehicle according to the video, and capturing a vehicle image from the video according to the driving track of the to-be-cleared vehicle; extracting positive and negative samples from the vehicle picture according to the spatial and temporal features of the vehicle, training a vehicle apparent feature extractor according to the samples, and extracting apparent features through the trained vehicle apparent feature extractor; comparing the apparent characteristics with other vehicles, obtaining an appearing road section of the to-be-sorted vehicle in the second road condition video according to a result, and simulating a possible driving path of the to-be-sorted vehicle; calculating an instantaneous speed according to the road condition video; determining a driving path according to the possible driving path and the instantaneous speed; and according to the driving path and a charging standard corresponding to the driving path, carrying out fee clearing calculation among the highway investment main bodies. By adopting the method, clearing of highway cost can be realized more accurately.

Description

technical field [0001] The invention relates to the technical field of road toll clearing, in particular to a method and device for expressway toll clearing. Background technique [0002] With the rapid development of my country's expressway network, the inter-provincial expressways are gradually forming a network. Due to the diversification of investment channels for expressway construction in my country, the setting of toll stations cannot completely coincide with the road sections constructed by investors. Therefore, it is extremely important to accurately classify expressway fees. [0003] At present, the domestic highway network fee clearing method discloses the use of deep convolutional network for image feature extraction to obtain the license plate number of the vehicle, and then uses the shortest path of the vehicle corresponding to the license plate number as the basis for vehicle billing. Or use the GPS positioning function of the mobile communication terminal an...

AI Technical Summary

Problems solved by technology

[0004] However, in the above-mentioned method for clearing highway network fees, the former will be affected by the clarity of license plates of vehicles on the highway, while the latter depends on the positioning accuracy of GPS and the data transmission speed of mobile terminals. These unstable factors will make the identified path Deviating from the real driving path of the vehicle, the purpose of accurately identifying the driving path of the vehicle cannot be achieved, which affects the income of expressway investors

Images