| With the great investment in the construction of traffic infrastructure in China,the construction of mountainous road has also developed rapidly.Although mountainous road provide convenience for people's life production,increase economic benefits,and promote social prosperity,it also brings serious traffic safety problems.As a result,it makes the requirements of the road users for the driving environment more and more higher.The landscape environment of mountainous roads is complex and changeable.When vehicles driving in this environment,drivers need to adjust the driving behavior in time to adapt to the changing driving environment.During this process,the driver's visual psychology changed,producing visual load.Then the safety of driving is affected.The mechanism of the influence of the landscape environment of mountainous roads on the driver's visual load is explored to provide a scientific and rational theoretical method for the construction of road landscapes in mountainous areas,and it can also ensure the safety of driving and provide theoretical support for the scientific construction of mountainous roads.This thesis is based on the Chongqing Science and Technology Commission's "Research on two-lane highway traffic safety in mountainous area based on visual psychology".On the basis of real environment driving experiment and related research results at home and abroad,the study of driver's visual load which was caused by mountainous road landscape is explored.The driver's visual information calculation model based on the mountainous road landscape was established.The relationship between sample entropy of driver's pupil area and driver's visual load intensity and driver's visual information in different landscape space types was analyzed.The visual safety evaluation system for mountainous road landscape based on driver's visual load was established,and the evaluation model was verified by specific examples in the last.First of all,the road of S105 Yujian to S415 Mulong in Chongqing was taked in this thesis as the experimental section to carry out driving experiments.The dynamic image of the road landscape and the driver's visual and psychological indicators were recorded in the experiment.The driver's visual psychological parameters are analyzed and screened,and the driver's pupil area is determined as an index to reflect driver's changing visual load.Second,on the basis of the analysis of previous researcher's results,the theory of visual power and calculation of visual load were applied in this thesis.The HSV color model is used to characterize the landscape environment,and the iteration method based on maximum threshold is used to segment the image of the visual field.And it also used to extract the color,saturation and brightness components of the image.Finally,the driver's visual information calculation model based on the mountainous road landscape was established,and the visual information generated by the mountainous road landscape environment for the driver was calculated through the model to achieve the quantification of mountainous road landscape environment.In addition,based on the theory of enclosure,this thesis divides the landscape environment of mountainous road into three types: open type,semi-closed type and closed type.Then,the driver's pupil area sample entropy is used to characterize the driver's visual psychological changes in this thesis,and the necessity and feasibility of using sample entropy characterization are analyzed and described in detail.And the relationships between sample entropy of driver's pupil area and driver's visual load intensity and driver's visual information in different landscape space types was analyzed in turn,and the relationship between them were obtained.In addition,the distribution of the three in different landscape space types were analyzed in turn,and the differences and relations between the three landscape space types were described.Finally,the driver's visual parameter information and driving environment information were analyzed.The visual safety evaluation system for mountainous road landscape based on driver's visual load was established.AHP and expert scoring method are used to determine the weight values of various levels of indicators in this thesis,and fuzzy comprehensive evaluation method is used to establish the visual safety assessment model of mountainous road landscape based on driver's visual load.After taking a section of road in Chongqing as an example and using the established evaluation model,the visual safety was evaluated comprehensively to obtaine the corresponding evaluation grade.And the reasons for the evaluation result were analyzed,and the corresponding improvements were proposed. |
PDF Full Text Request