Method and device for realizing classification processing, computer storage medium and terminal

Abstract

The invention discloses a method and a device for realizing classification processing, a computer storage medium and a terminal. The method comprises the steps of determining a vehicle type of a to-be-analyzed vehicle based on a preset classification model; determining a vehicle model of the to-be-analyzed vehicle by adopting a regression model corresponding to the vehicle type of the to-be-analyzed vehicle; wherein the vehicle types comprise a passenger car and a truck; wherein the vehicle model comprises a preset classification number for distinguishing the vehicle toll. According to the embodiment of the invention, the regression model is adopted to determine the vehicle model after the vehicle type is determined, so that accurate recognition of the vehicle is realized, and technical support is provided for assisting an electronic toll collection (ETC) system to obtain accurate vehicle information.

Description

technical field [0001] This article involves but is not limited to the realization of image processing technology, especially a method, device, computer storage medium and terminal for realizing classification processing. Background technique [0002] Toll roads refer to roads where tolls are levied on passing vehicles in order to repay loans or raise funds. There will be different charging standards for different vehicles, such as cars with less than seven seats, cars with more than seven seats, passenger buses, large buses and other passenger vehicles, low-load trucks, high-load trucks and other trucks will have different charging standards; The larger the passenger car, the higher the toll charged; the larger the truck, the higher the toll charged; the toll charges for passenger cars and trucks are different. Electronic Toll Collection (ETC, Electronic Toll Collection) system is an automated system used on highways or bridges to collect tolls for passing vehicles. ETC id...

AI Technical Summary

Problems solved by technology

The regression model, like the classification model, needs a large amount of manually labeled training data for model training, so it is still difficult to implement; in addition, the regression model cannot be used to obtain vehicle types; although different types of vehicles correspond to different vehicle sizes, trucks and The difference between a passenger car is not the same as the difference between a truck and a truck. Using a regression model for vehicle type judgment may misjudge a passenger car as a truck and a truck as a passenger car; therefore, the regression model cannot be used to identify vehicle types

Images