本文選題：視覺(jué)注意轉(zhuǎn)移 + 聚類分析�。� 參考：《長(zhǎng)安大學(xué)》2015年碩士論文

【摘要】：近年來(lái),我國(guó)高速公路發(fā)展迅速,與之相對(duì)應(yīng)的交通事故也是層出不窮。在道路交通安全的領(lǐng)域中,隧道路段的安全狀況擁有獨(dú)特和重要的地位。在隧道路段的駕駛過(guò)程中,駕駛?cè)说囊曈X(jué)注視轉(zhuǎn)移與普通道路有著很大的差異性。本文旨在研究駕駛?cè)嗽隈{駛車輛經(jīng)過(guò)隧道過(guò)程中,視覺(jué)的注視轉(zhuǎn)移規(guī)律,希望研究?jī)?nèi)容對(duì)提高隧道路段安全和對(duì)駕駛員視覺(jué)研究能夠有所裨益。本研究在真實(shí)環(huán)境下進(jìn)行了實(shí)車試驗(yàn),選取包茂高速陜西段的9個(gè)隧道道路作為試驗(yàn)線路,選取三名駕駛?cè)?采用SR Research公司出產(chǎn)的Eye Link II型眼動(dòng)儀追蹤記錄駕駛?cè)说难蹌?dòng)行為特征。同時(shí),選用GPS和五輪儀采集坐標(biāo)、位移和速度等相關(guān)數(shù)據(jù)。在對(duì)國(guó)內(nèi)外研究進(jìn)行回顧和整理的基礎(chǔ)上,根據(jù)實(shí)際情況和數(shù)據(jù)的聚類分析對(duì)隧道路段的駕駛?cè)说囊曈X(jué)注視分布區(qū)域進(jìn)行劃分,共八個(gè)區(qū)域,分別為:前面遠(yuǎn)方、車內(nèi)后視鏡、左方、正前方、右方、左后視鏡、右后視鏡、右方、駕駛儀器。基于所采集的數(shù)據(jù),結(jié)合試驗(yàn)路線特點(diǎn),分析了駕駛?cè)嗽谒淼赖倪M(jìn)口段、中間段和出口段的視覺(jué)轉(zhuǎn)角轉(zhuǎn)移頻次和視覺(jué)分布情況,運(yùn)用馬爾可夫鏈理論分析了注視點(diǎn)在各區(qū)域之間的一步和兩步轉(zhuǎn)移規(guī)律。研究結(jié)果表明,駕駛?cè)说囊曈X(jué)注意轉(zhuǎn)移頻次在隧道進(jìn)口段和隧道出口段中,在隧道口有一個(gè)短暫的快速上升。在隧道里駕駛?cè)说囊曈X(jué)頻次略低于在普通路段。駕駛?cè)说囊曈X(jué)注意轉(zhuǎn)移注視分布大概是正前方前面遠(yuǎn)方左(右)方左(右)后視鏡駕駛儀器車內(nèi)后視鏡。在進(jìn)口段和出口段駕駛?cè)说淖蠓胶陀曳降囊曈X(jué)分布上升,在隧道中間段的駕駛?cè)饲懊孢h(yuǎn)方的視覺(jué)分布很少。同普通路段相比,隧道路段的正前方的視覺(jué)分布略低于普通路段。駕駛?cè)说囊曈X(jué)注意轉(zhuǎn)移概率主要與三個(gè)因素有關(guān):一、當(dāng)前所在區(qū)域,最大的轉(zhuǎn)移概率一般是停留在自身區(qū)域;二、視覺(jué)分布的多少,正前方的視覺(jué)分布最多,所以當(dāng)前區(qū)域往正前方的視覺(jué)注意轉(zhuǎn)移概率一般較大;三、空間位置的分布,一般來(lái)說(shuō),跳過(guò)某區(qū)域,從當(dāng)前區(qū)域跳到空間位置很遠(yuǎn)區(qū)域的概率一般較小,轉(zhuǎn)移到周圍區(qū)域的概率更大。本論文得到國(guó)家自然科學(xué)基金項(xiàng)目《駕駛員視覺(jué)注意聚焦及轉(zhuǎn)移模式與隧道空間通視性研究》(項(xiàng)目批準(zhǔn)號(hào):51308059)的支持。
[Abstract]:In recent years, our country highway develops rapidly, the corresponding traffic accident also appears endlessly.In the field of road traffic safety, tunnel safety has a unique and important position.In the driving process of tunnel section, there is a great difference between the visual gaze shift of the driver and the ordinary road.The purpose of this paper is to study the rules of visual gaze transfer in the course of driving through a tunnel, and hope that the research can be helpful to improve the safety of tunnel section and to study driver's vision.In this study, a real vehicle test was carried out in real environment. Nine tunnel roads in Baomao Expressway Shaanxi Section were selected as the test lines, and three drivers were selected.The Eye Link II eye movement instrument produced by SR Research Company was used to track and record the eye movement behavior of the driver.At the same time, GPS and five-wheel instrument are used to collect coordinate, displacement and velocity.Based on the review and arrangement of the domestic and foreign studies, according to the actual situation and the clustering analysis of the data, the visual gaze distribution area of the driver in the tunnel section is divided into eight regions, which are: the front far away, the rear view mirror inside the vehicle.Left, front, right, left rearview mirror, right rear view mirror, right side, driving instrument.Based on the collected data and the characteristics of the test route, the frequency and distribution of the visual angle transfer of the driver in the inlet, middle and outlet sections of the tunnel are analyzed.Based on Markov chain theory, the one-step and two-step transfer laws of fixation points in different regions are analyzed.The results show that the shift frequency of the driver's visual attention in the entrance and exit of the tunnel has a transient and rapid rise at the entrance of the tunnel.The visual frequency of the driver in the tunnel is slightly lower than that in the normal section.The driver's visual attention shift focus distribution is probably the front of the far left (right) square left (right) rearview mirror driving instrument vehicle rearview mirror.The visual distribution of the left and right side of the driver in the inlet and outlet sections is increased, while that of the driver in the middle section of the tunnel is less than that of the driver in front of the middle section of the tunnel.Compared with the ordinary section, the visual distribution of the front of the tunnel section is slightly lower than that of the ordinary section.The probability of the driver's visual attention transfer is mainly related to three factors: first, the maximum transfer probability of the current region is generally staying in its own region; second, the number of visual distribution, the visual distribution in front of the most,Therefore, the probability of visual attention transfer from the current region to the front of the region is generally high. Thirdly, the distribution of the spatial position, generally speaking, skipping a certain area, jumping from the current region to the far area of the space, is generally less likely.The probability of moving to the surrounding area is greater.This thesis is supported by the National Natural Science Foundation of China (project Grant No.: 51308059).
【學(xué)位授予單位】：長(zhǎng)安大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類號(hào)】：U491.25

【參考文獻(xiàn)】

相關(guān)期刊論文 前2條

1 張延安;試論馬爾可夫模型及應(yīng)用[J];沈陽(yáng)大學(xué)學(xué)報(bào);2001年02期

2 戴憂華;郭忠印;;空間通視性對(duì)高速公路隧道路段駕駛行為的影響[J];同濟(jì)大學(xué)學(xué)報(bào)(自然科學(xué)版);2011年09期

，

本文編號(hào)：1731501