论文部分内容阅读
为研究黄昏时段不同颜色引起驾驶员空间距离判识差异的变化规律,选用32名驾驶员,在黄昏环境照度变化的实际道路中,对不同深度距离下红、绿色障碍物的空间绝对距离和相对距离进行判识。比较人在三维空间中对红、绿色障碍物距离判识结果,获得判识距离差异特征值。对所获得的试验结果,运用BP神经网络模拟距离判识差异随照度和深度距离不同而变化的规律。结果表明:在三维空间,红、绿色障碍物黄昏时段距离判识差异显著,绿色障碍物判识距离大于红色,BP网络可以很好地拟合距离判识差异变化规律;随着外界环境照度的下降,红、绿颜色引起的距离判识值差异均增大,其中绝对距离差异增加明显;随着空间深度方向距离增加,判识差异也增大,其中相对距离判识差异增加量较绝对距离小。
In order to study the variation of driver’s spatial distance discrepancy caused by different colors at dusk, 32 drivers were selected to determine the spatial absolute distance between red and green obstacles at different depth distances and relative Distance to judge. Comparison of the distance between red and green obstacles in three-dimensional space results in identification of distance differences to obtain the eigenvalue. The obtained test results, the use of BP neural network simulation distance differences with the illumination and depth of the distance varies. The results show that in three-dimensional space, the distance between red and green obstructions is significantly different at dusk, and the distance between green obstructions is greater than that of red. The BP network can well fit the variation of distance discretization. The differences between the distance, the red, and the green color caused by the distance discrimination value all increased, and the absolute distance difference increased obviously. As the distance in the spatial depth direction increased, the discrimination difference also increased, in which the relative distance difference increased more than the absolute distance small.