Fatigue driving behavior detection system based on parallel cross convolutional neural network

Abstract

The invention relates to a fatigue driving detection system based on a parallel cross convolutional neural network, and the system comprises the steps: collecting a fatigue driving sample image, dividing the sample image, constructing a parallel cross convolutional neural network model to train a collected driving state sample, and enabling a model obtained through training to be used for fatiguedriving detection. The parallel cross convolution neural network model can effectively extract image features and meet the real-time requirement. Meanwhile, the invention provides a fatigue driving detection system function assembly based on the parallel cross convolutional neural network, the fatigue driving detection system function assembly comprises a camera, a voice prompt module, a timer andthe like, the operation process of the fatigue driving detection system is described in detail, the system can monitor the state of a driver in real time, and accidents are effectively prevented.

Description

technical field [0001] The invention relates to a fatigue driving behavior detection system based on a convolutional neural network, which belongs to the technical field of automobile safety driving. Background technique [0002] With the rapid development of society, people's travel methods have become richer, and cars have become the most common means of transportation and means of transportation in our lives; are also occurring more and more frequently. According to statistics, there are at least 600,000 deaths caused by traffic accidents worldwide every year; and about 90% of traffic accidents are caused by drivers, most of which are fatigue driving. If it is possible to monitor whether the driver is in a state of fatigue driving, and to call the police immediately once it is found, the incidence of such traffic accidents can be greatly reduced. Therefore, it is of great significance to develop a fatigue driving behavior detection system. [0003] Fatigue driving refe...

AI Technical Summary

Problems solved by technology

[0004] The detection method based on the vehicle state (trajectory change and lane deviation) is to install various sensors that can detect the vehicle state, and analyze the driver's behavior to judge the fatigue driving situation through the vehicle state. This method has high hardware requirements and is expensive. , and is also easily affected by the driving environment and driver's habits, and the error is large

[0005] The accuracy of judging fatigue driving based on physiological signals (EEG signals, ECG signals, etc.) is high, but this acquisition method requires the driver to wear some sensors, which is complicated to operate and inconvenient to wear, which brings inconvenience and limitations to practical applications

[0006] The detection method based on the driver's physiological response characteristics (eye characteristics, mouth movement and other facial expressions) is currently mainly installed in the car with a camera, and the traditional machine learning method is used for detection and analysis. Compared with the previous two methods, this method is interfered. The factor is small, easy to use, and cheap, but it is easily affected by factors such as illumination and driver's posture, resulting in low accuracy

Images