On the plausible association between environmental conditions and human eye damage

The increase in solar ultraviolet radiation can have various direct and indirect effects on human health, like the incidence of ocular damage. Data of eye damage in residents of three suburban regions in Greece and in two groups of monks/nuns and fishermen are examined here. The statistics performed on these data provides new information about the plausible association between increased levels of solar ultraviolet radiation, air-pollution at ground level, and the development of ocular defects.     

基于视线追踪技术的工艺操作人员人为失误识别研究

为了对生产加工过程中工艺操作人员的不安全行为进行有效监控，及时避免因操作失误带来的工艺安全事故，开展基于视线追踪技术的工艺操作人员人为失误识别研究。通过设计眼动实验方案，利用视线追踪技术采集操作者在工艺流程控制过程中各类失误模式的眼动数据,并对采集的数据进行统计，提取失误行为的眼动特征，并建立人为失误智能识别方法，利用距离函数聚类实现现场操作人员失误状态的准确识别。研究结果表明:以被试样本在不同区域的视线停留时间百分比作为特征参数，并运用欧氏距离函数分类法判别人员操作状态，对操作人员的不安全行为能够准确识别。     

安全标志排列顺序的视觉注意特征研究*

为研究安全标志位置设置的视觉注意特征,选取国标中4种安全标志（警告、禁止、指令、提示）进行排列组合,利用Eyelink Ⅱ记录20名被试的平均注视时间和首次注视时间,并采用统计学及视觉注意理论进行分析。研究结果表明:从位置角度分析,无论横向还是纵向排列,位置Ⅱ的首次注视时间相对较短,且当禁止标志位于位置Ⅱ时,首次注视时间平均值相对较小;从平均注视时间角度分析,禁止标志位于位置Ⅱ时,平均注视时间相对较长;通过横纵位置Ⅱ的对比可知,禁止标志位于位置Ⅱ时,排列方式对平均注视时间及首次注视时间无显著影响（p>0.05）;横向中偏左的位置及纵向中偏上的位置具有较好的识别性,而禁止标志位于位置Ⅱ时,显著性更高且横向与纵向无差异,国标的顺序设置更能增强安全标志的显著性。     

Work Content and Eye Discomfort in VDT Work

A questionnaire study of Swedish public employees with frequent and different types of VDT work (n = 2,025) was conducted. Eight different kinds of eye discomfort were examined, and an additive index was constructed. Eye discomfort was statistically related to aspects of work organization. The highest levels of discomfort were reported by computer-aided design (CAD), data entry, and word processing groups; groups with mixed VDT tasks displayed the lowest level of symptoms. Time spent at a VDT, low degree of work control, time pressure, and high pace of work were all associated with relatively high levels of eye discomfort. There was significant interaction between experience of stress and time spent at a VDT with respect to both the eye discomfort index and three specific symptoms (itching, gritty feeling, and dryness). Whereas dryness, smarting, and itching of the eyes seem most strongly related to work conditions, redness and watery eyes seem most weakly related.     

基于视线追踪的三维模拟灭火救援培训效果评估

本文结合灭火救援指挥三维计算机模拟训练平台以及桌面式视线追踪实验系统,开展基于视线追踪技术的学员应急培训效果评估。首先开展眼动分析实验系统搭建与实验设计,采用Eyeso Ec80遥测式眼动仪对操作者模拟灭火救援时的眼动信息进行数据采集与处理。然后,开展失误行为的眼动特征提取研究,针对不同类别的认知状态,分别统计其在不同兴趣区域的注视时间并建立一步转移概率矩阵,最终得到典型失误模式的失误特征。     

Physical and motor development in male and female rat offspring prenatally exposed to gamma radiation

Abstract

The effects of radiation on physical and motor development of male and female rats exposed to ionizing radiation in utero were studied. Rats were exposed to three different doses of radiation (150 rad, 15 rad and 6.8 rad, considered as high, moderate and low doses of radiation respectively) on the 20th day of prenatal life. Exposure to 150 rad contributed to significantly lower body weights of both male and female rat offspring. Upper jaw tooth eruption was delayed in 150 rad treated male offspring, as well as in 15 rad and 150 rad treated female offspring. Cliff‐avoidance response was delayed in 6.8 rad, 15 rad, and 150 rad treated male offspring; and 150 rad treated female offspring. Lower jaw tooth eruption, eye opening, and crawling were not affected by radiation in male or female animals. Results indicate that radiation affected the developmental parameters of both male and female rat offspring, and sex of the offspring played no role in the magnitude of radiation induced damages.     

基于面部特征识别的管制员疲劳监测方法研究

为进一步控制疲劳诱发的空管人为差错,提出一套基于面部特征识别的空中交通管制员疲劳实时监测方法。应用眼动仪和视频记录系统开展36 h睡眠剥夺试验,确定PERCLOS值、平均闭眼时长、哈欠频率3个疲劳判定指标的判定阈值,并给出综合疲劳程度的融合算法。应用机器视觉的理论和方法,开发功能完整的管制员疲劳状态实时监测告警原型系统。结果表明,通过对面部特征的识别和状态融合能够实现对管制员疲劳状态的实时监测。建议在此基础上进一步开发可实用的疲劳监测系统。     

基于多模态信息融合的心理负荷评估*

为了相对稳定地识别人员是否处于心理负荷状态,设计并实施含能材料起爆作业心理负荷诱导试验。首先对心理负荷诱导情况进行判断,然后分别采集27名被试在静息状态和实施含能材料撞击起爆作业状态下的眼动、心率变异性（HRV）和脑电信号（EEG）,通过正态性检验和假设检验获得心理负荷表征指标并进行统计功效分析,依据表征指标,采用支持向量机（SVM）和随机森林（RF）算法建立多模态信息融合的心理负荷评估模型,最后采用被试工作特征曲线（ROC曲线）分析各模态组合和分类器的心理负荷识别性能。研究结果表明:双模态（眼动+EEG）下SVM算法和3模态下RF算法评估性能和稳健性较高,多模态信息组合具有优异的识别效果。