Vehicle type identification method and system based on remote supervised learning

Abstract

The invention discloses a vehicle type identification method and system based on remote supervised learning, and belongs to the technical field of intelligent traffic, and the method comprises the following steps: S1, obtaining a road picture transmitted by a road monitoring camera, and according to the position of a to-be-identified vehicle in the road picture, determining the position of the to-be-identified vehicle in the road picture; obtaining a vehicle body picture of a to-be-recognized vehicle from the road picture; and S2, preprocessing the vehicle body picture of the vehicle, then performing vehicle type identification according to the vehicle body picture of the to-be-identified vehicle, and obtaining vehicle type identification information of the to-be-identified vehicle. The method has the advantages of being high in recognition accuracy and fast in training convergence, the calculation speed can meet the processing requirement of a smart road for vehicle type recognition, the method can be effectively applied to smart traffic information, the method further has the remote supervised learning ability, the vehicle type recognition information base can be automatically updated, the vehicle type recognition ability is higher, and the vehicle type recognition efficiency is improved. The method can well adapt to accurate identification of vehicle types in various environments.

Description

technical field [0001] The invention belongs to the technical field of intelligent transportation, and in particular relates to a vehicle type recognition method and system based on remote supervision learning. Background technique [0002] Vehicle type recognition is an important research content in the field of intelligent transportation systems. With the acceleration of the pace of life in modern society, automobiles have increasingly become an important means of transportation in people's lives and work. The widespread use of automobiles has provided convenience for people's travel, but it has greatly increased the traffic operation load. [0003] With the fast-paced development of society, the video image data that the traffic management department needs to process every day has also increased sharply, and the appearance of automobiles and vehicles has been updated more frequently, which is tantamount to further increasing the work intensity of the traffic management d...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is: in order to solve the problem that with the fast-paced development of society, the video image data that the traffic management department needs to process every day also increases sharply, and the appearance update speed of automobiles and vehicles is also more frequent, which is tantamount to further increasing traffic management. The work intensity of the department, and the existing vehicle type identification method has poor generalization ability and does not have the self-learning function. When encountering a new type of vehicle, the vehicle type identification often fails due to lack of information, and then it is transferred to manual To identify channels, low efficiency, and low recognition accuracy, a vehicle type recognition method and system based on remote supervision learning is proposed