可视化技术及其在铁路微机联锁系统中的应用

作者结合我国铁路人员的现状,研究提出采用可视化和人工智能技术开发微机联锁系统的方法,从而可在设计、开发和使用维护微机联锁系统各个阶段提高系统的可靠性和安全性。     

提高级配指标计算精度方法研究

级配曲线是颗粒分析试验的重要成果图件,用其计算的不均匀系数Cu与曲率系数Cc是路基填料划分的重要指标。传统作图法整理级配曲线时,工作量大且精度不高。在分析线性插值、多项式插值、Lagrange插值与分段抛物线插值算法的基础上,提出了修正样条函数拟合的方法,得到土的光滑颗粒级配曲线。针对非单调样条函数,采用最佳逼近算法求解,克服了曲线出现平台时病态函数问题。文中提供了求解的关键代码,实践表明该法可以提高计算级配指标的精度和自动化程度。     

基于离散小波变换和人类视觉系统的数字水印算法

通过对人类视觉系统(HVS)的亮度特性、频域特性、图像类型特性的研究,提出一种基于离散小波变换和人类视觉系统数字水印算法,将秘密信息嵌入到小波变换不同的方位系数中.实现了水印信号在图像信息中的嵌入与提取,达到隐藏秘密信息的目的.实践证明该算法能提高水印的不可见性,同时具有很强的鲁棒性.     

The “field of safe travel” revisited: interpreting driving behaviour performance through a holistic approach

The present paper proposes a conceptual framework for the driver’s visual–spatial perceptual processes. Based on a theoretical analysis of driving proposed by Gibson and Crooks [(1938). A theoretical field-analysis of automobile-driving. The American Journal of Psychology, 51, 453–471. doi:10.2307/1416145], the developed field of safe travel (FoST) framework suggests that at any moment the driver constructs a “field” by integrating two perceptual entities: (i) the possible available spatial fields for locomotion and (ii) the driver’s mental image of ego-vehicle outer-line and motion dynamics. This framework is used to reinterpret in a unified way a number of disparate research findings reported in the literature concerning specific driving sub-tasks (e.g. lane keeping and car following). It is argued that the FoST framework may be used to predict drivers’ behaviour in various traffic/situation environments based on their prioritisation between the above two perceptual entities. Implications of the proposed framework at a theoretical and practical level, in view of the future of driving with multiple levels of automation, are also discussed.     

手机社交娱乐操作对驾驶人视觉参数影响研究

为探究手机社交娱乐操作对驾驶人视觉参数的影响,基于模拟驾驶器和SmartEye 眼动仪设计模拟驾驶实验,采集驾驶人正常驾驶、拨打电话、收发信息等5种操作下视觉特征数据,应用拉依达准则和联合频数分布直方图进行数据归约.针对不同操作下视觉参数,应用箱形图和Kruskal-Wallis显著性检验,验证视觉参数评价指标的有效性.结果表明：随手机操作类型的不同,驾驶人水平和垂直视角标准差、视线离开道路时间百分比、扫视平均速度和眨眼持续时间呈显著性变化,可作为视觉参数影响的评价指标;观看视频和刷朋友圈操作对驾驶人视觉影响程度明显高于拨打电话和收发短信操作.     

山区公路驾驶视觉信息量计算方法研究

为解决山区公路中驾驶视觉信息量难以量化的问题,对驾驶视野图像进行分割,根 据HSV颜色模型,提取视野图像的色调、饱和度、亮度值,再结合车速值,在驾驶视觉心理负荷的基础上,提出山区公路路域环境下的驾驶视觉信息量计算方法.通过实车实验,进行数据采集,并验证计算方法.计算结果表明,在半郁闭型空间行驶时,接收的视觉信息量最大;在郁闭型空间中,接收的信息量最小.计算结果与被试实际感受具有一致性,说明本文提出的驾驶视觉信息量计算方法具有可行性,可为路域环境的合理布设提供一定的技术参考.     

隧道口减光设施对照明的影响研究

近年来,伴随着我国基础设施建设的不断推进,隧道里程不断增多,隧道照明系统的重要性日益突出,因此对隧道洞口驾驶员视觉特性、减光设施的特点及减光效果进行研究,具有重要的意义。本文通过对隧道洞口驾驶员的视觉特性进行研究,分别论述各种洞口减光设施的特点,并以南京扬子江路隧道、富贵山隧道、双龙街立交隧道的检测数据为基础,计算减光设施对洞口亮度的影响,以及隧道内加强照明的亮度需求的降低程度,进一步分析得出采用减光设施对照明设施建设的投资影响以及可节约的照明养护成本等。     

Real time detection of driver attention: Emerging solutions based on robust iconic classifiers and dictionary of poses

Real time monitoring of driver attention by computer vision techniques is a key issue in the development of advanced driver assistance systems. While past work mostly focused on structured feature-based approaches, characterized by high computational requirements, emerging technologies based on iconic classifiers recently proved to be good candidates for the implementation of accurate and real-time solutions, characterized by simplicity and automatic fast training stages.In this work the combined use of binary classifiers and iconic data reduction, based on Sanger neural networks, is proposed, detailing critical aspects related to the application of this approach to the specific problem of driving assistance. In particular it is investigated the possibility of a simplified learning stage, based on a small dictionary of poses, that makes the system almost independent from the actual user.On-board experiments demonstrate the effectiveness of the approach, even in case of noise and adverse light conditions. Moreover the system proved unexpected robustness to various categories of users, including people with beard and eyeglasses. Temporal integration of classification results, together with a partial distinction among visual distraction and fatigue effects, make the proposed technology an excellent candidate for the exploration of adaptive and user-centered applications in the automotive field.     

The speed control effect of highway tunnel sidewall markings based on color and temporal frequency

The low‐luminance monotonous environment in the middle section of highway tunnels offers few reference points and is prone to cause severe visual illusion. Thus, drivers tend to underestimate their driving speed, which can induce speeding behaviors that result in rear‐end collisions. Therefore, discovering low‐cost methods of traffic engineering that reduce this visual illusion and ensure a steady driving speed is an important challenge for current highway tunnel operations. This study analyzes the effects of sidewall markings in typical highway tunnels, specifically observing how their colors and temporal frequencies affect the driver's speed perception in a low‐luminance condition. A three‐dimensional model of the middle section of highway tunnels was built in a driving simulator. Psychophysical tests of speed perception were carried out by the method of limits. The precision of the simulation model was then checked by comparing the results to field test data. The simulation tests studied the stimulus of subjectively equal speed and reaction time in relation to sidewall markings in different colors (red–white combined, yellow–white combined, and blue–white combined). Furthermore, based on the optimal color, the effects of sidewall marking with different temporal frequencies (0.4, 0.8, 1.2, 2, 4, 8, 12, 16, and 32 Hz) on the speed perception of drivers were also analyzed. The test results reveal that the color and temporal frequency of sidewall marking have a significant impact on the driver's stimulus of subjectively equal speed and reaction time. The subjects have the highest speed overestimation and an easy speed judgment with the red–white combined sidewall marking. Within the temporal frequency range of 4.45–7.01 Hz, the subjects have a certain degree of speed overestimation (less than 20%), and the speed perception is sensitive to the temporal frequency changes. Copyright © 2016 John Wiley & Sons, Ltd.     

基于人-车交互的行人轨迹预测

针对行人轨迹预测具有复杂、拥挤的场景和社会交互问题,基于长短时记忆网络(Long Short-term Memory Network,LSTM)对行人与车辆、行人与其他行人的交互进行建模,提出一种基于人-车交互的行人轨迹预测模型(VP-LSTM).该模型同时考虑了行人与行人的交互、行人与车辆的交互,更适用于复杂的交通场...