A driving behavior recognition system based on convolution neural network

Abstract

The invention discloses a building method of a driving behavior recognition system based on a convolution neural network. The following steps are sequentially carried out: collecting driving behaviorsample data; data filtering; data format regularization; driving behavior recognition takes the regularized driving behavior sample data as input and inputs it to the constructed convolution neural network, and outputs it as the driving behavior type to be recognized after pooling. Driving behavior is trained by convolution neural network. The invention solves the problem that the traditional machine relies on manual selection when recognizing driving behavior in the field of driving behavior recognition, the accuracy of recognition is not high, at the same time. The invention provides a method for building a driving behavior recognition system based on a convolution neural network, which is used for automatically extracting features according to the system needs to improve the recognitionaccuracy of driving behavior, and can effectively utilize a large-scale driving behavior data set to recognize more kinds of driving behavior.

Description

technical field [0001] The invention relates to the field of deep learning, in particular to a method for building a driving behavior recognition system based on a convolutional neural network. Background technique [0002] With the rapid development of economy and society, the number of motor vehicles in my country is increasing rapidly. The rapid increase in the number of motor vehicles has brought convenience to people's travel, but it has also led to increasingly serious traffic congestion and frequent traffic accidents. According to statistics from the Traffic Management Bureau of the Ministry of Public Security, 90% of fatal traffic accidents are caused by human factors. Based on the above considerations, the present invention recognizes human driving behaviors, such as acceleration, braking, turning, lane changing, continuous lane changing, and overtaking, etc., and feeds them back to drivers or relevant traffic control departments to remind drivers to drive carefull...

AI Technical Summary

Problems solved by technology

[0005] The invention solves the problem that traditional machines rely on manual selection when identifying driving behaviors in the field of driving behavior recognition, the recognition accuracy is not high, and at the same time, traditional machines cannot handle large-scale data, and provide a driving behavior recognition system based on convolutional neural networks. The method, which is based on the convolutional neural network to automatically extract features according to the needs of the system, improves the recognition accuracy of driving behaviors, and can effectively use large-scale driving behavior data sets to identify more types of driving behaviors

Images