Vehicle model identification method, device and equipment based on convolutional neural network

Abstract

The invention discloses a vehicle model identification method, device and equipment based on a convolutional neural network. The vehicle model identification method based on the convolutional neural network specifically comprises the following steps that: utilizing a first quantity of convolutional layer units to extract the local features of a vehicle image to be identified; on the basis of the local feature, utilizing a second quantity of convolutional layer units to extract the global features of the vehicle image to be identified; and according to the local features and the global features, utilizing a classification layer to identify the vehicle model of a vehicle in the vehicle image. By use of the method, the local features and the global features extracted by the convolutional layer units in the convolutional neural network are input into the classification layer, the local features and the global features of the vehicle image to be identified can be both given a considerationin vehicle modal identification, the problem that accuracy is lowered since feature detail information is lost because image features are too single is avoided, and the accuracy of an identification result is improved.

Description

technical field [0001] The present application relates to the technical field of machine vision, and in particular to a method, device and equipment for vehicle identification based on convolutional neural network. Background technique [0002] With the construction of cities and the development of society, the number of vehicles on the streets is increasing, and the road conditions are becoming more and more complex. Traffic management is facing challenges in many aspects-vehicle congestion, traffic accidents, road obstacles, etc. It is far from enough to simply rely on the formulation of relevant regulations and manual monitoring by relevant departments, so a practical and effective solution is needed. Building an intelligent transportation system is an effective method, and it is also the trend of urban transportation development. [0003] The detection and identification of vehicles is the technical core of the intelligent transportation system. Vehicle identification p...

AI Technical Summary

Problems solved by technology

[0005] However, when using the existing technology to identify car models, not only does it need to pre-train multiple convolutional neural networks, but also causes errors in the image segmentation process, which affects the accuracy of the recognition results

Images