通过行为实验,采集了两组被试在不同注意状态下观看道路交通场景图像时的眼动数据,研究了在交通驾驶环境中两种不同注意机制(自底向上和自上而下)驱动下的眼动特征。研究表明,在交通驾驶环境中,基于自底向上注意驱动的眼动特征与基于自上而下注意驱动的眼动特征存在着显著性差异,二者对应的视觉显著图也有明显区别。另外,两种注意机制对信号灯及交通标识等重要交通元素的关注及识别等方面也存在显著的不同。
Human visual system is a complicated information processing system. Selective attention is an important mechanism which enables us to process relevant inputs from a large amount of visual information. Traffic environment is a complex and tridimensional scene of multiple information sources, which changes dynamically and requires being processed instantly. The driver's attention is always controlled by two visual attention mechanisms, namely bottom-up and top-down attention. The bottom-up attention is driven by the environment and image features, and top-down attention is based on tasks and cognitive experiences. In this paper, the behavioral experiment was carried out to investigate the features of eye movements under these two selective attention mechanisms in traffic environmen4 and also to acquire the dataset of saliency maps of real eye movement. Our results show that there are significant differences between bottom-up and top-down selective attention in saliency maps, characteristics of eye movement, and viewing of traffic signals and traffic signs, and so on.