A Bad Driving State Recognition Method Based on Multi-feature Convolutional Neural Network

Abstract

The invention discloses a bad driving state recognition method based on a multi-feature convolutional neural network, comprising: collecting inertial sensor data of a vehicle-mounted smart phone, performing preprocessing, and obtaining a source data set; dividing the source data set into individual data units , perform statistical feature extraction on each data unit, and label it to make a data set, named feature data set; build a multi-feature convolutional neural network, select appropriate network parameters and optimizers, and use the source data set and features The data set fully trains the multi-feature convolutional neural network, and obtains the trained multi-feature convolutional neural network model; uses the trained multi-feature convolutional neural network model to classify the inertial sensor data of the vehicle mobile phone, so as to realize the current Recognition of driving state, judging whether the current driving state of the car is a bad driving state, and recording and processing data in the background. The invention has the advantages of fast operation speed, high recognition rate and strong anti-environment interference ability.

Description

technical field [0001] The invention relates to the technical field of sensor data acquisition and deep learning, in particular to a method for identifying bad driving states based on a multi-feature convolutional neural network. Background technique [0002] With the rapid development of the automobile industry and the increasing popularity of automobiles, automobiles have become the most important means of transportation. However, some drivers still have the problem of irregular driving. The traffic control department and some online car-hailing platforms hope to monitor the driving status of drivers to evaluate their driving habits. [0003] At present, there are mainly three methods for detecting bad driving conditions. One is to detect dangerous driving conditions by installing different types of sensors or on-board computer systems on the car to reduce driving risks. The second is to judge whether the driver's driving condition is good or not based on the driver's ext...

AI Technical Summary

Problems solved by technology

At present, this type of method is mainly based on instantaneously collected data and uses sensor data change thresholds and traditional machine learning algorithms for analysis. The robustness and accuracy of the system need to be improved.

Images