Study On Safety Assessment Method Of Mountain Tourist Highway Based On Driver's Visual Characteristics

In recent years,with the national economic growth and the improvement of national living conditions,people's concept of tourism has gradually developed from traditional sightseeing tourism to leisure vacation tourism and experiential tourism.In December 2016,the State Council proposed that 25 national tourist scenic roads should be built throughout the country.More than 60% of these 25 national tourist scenic roads are located in mountainous areas.It is necessary to evaluate the safety of tourist highway in mountainous area.The research shows that 90% of road safety accidents are related to human beings.Based on the driver's visual characteristics,the safety evaluation of mountain tourist highway is of great significance.It also provides a new idea for the safety evaluation of mountain tourist highway in the future.In this paper,Smart Eye Pro 5.7 Eye Tracker is used to test the driver's dynamic vision in the mountainous tourist highway of Chongqing.The different road alignment,roadbed cross-section,signs and landscape types of the mountainous tourist highway are studied respectively,and the driver's visual characteristic parameters are obtained.SPSS,Excel and other software were used to analyze and process the data by correlation test and normal distribution test.Through the analysis of road alignment,roadbed cross-section,signs and landscape of mountain tourism highway by collecting driver's visual characteristics data,the influence of various indicators of mountain tourism highway on driver's visual characteristics(gaze time,pupil diameter,blink duration)is obtained,and the influence degree of different factors of mountain tourism highway on driving safety is analyzed.On the basis of driver's visual characteristics,the safety evaluation index of mountain tourism highway is put forward.The BP neural network framework model is used to obtain the indexes of pupil diameter,maximum change rate of pupil diameter and blink duration,and the classification and evaluation criteria of each index are elaborated in detail(some of them are classified according to three principles).The weight model of safety index of mountain tourism highway based on driver's visual characteristics is established by BP neural network to obtain the weight of each index.After analyzing the characteristics of each comprehensive evaluation method and combining with the characteristics of the mountain tourism highway safety evaluation system,this paper establishes a combination evaluation model of mountain tourism highway safety based on deviation value.Finally,taking a mountainous tourist highway in Chongqing as an example,this paper uses the established evaluation model to evaluate its safety,obtains the corresponding evaluation grade,and analyzes the reasons for the evaluation results.According to the actual analysis,it can be seen that the evaluation method basically coincides with the actual situation.This paper establishes an evaluation model which can be used as a means of the actual mountainous tourist highway safety evaluation.It can provide theoretical basis for the improvement of tourist highway in mountainous areas.