The wildfire prediction model is crucial for accurate rescue and rapid evacuation. Existing models mainly adopt regular grids or fire perimeters to describe the wildfire landscape. However, these models have difficulty in explicitly demonstrating the local spread details, especially in a complex landscape. In this paper, we propose a wildfire spread model with an irregular graph network (IGN). This model implemented an IGN generation algorithm to characterize the wildland landscape with a variable scale, adaptively encoding complex regions with dense nodes and simple regions with sparse nodes. Then, a deep learning-based spread model is designed to calculate the spread duration of each graph edge under variable environmental conditions. Comparative experiments between the IGN model and widely used fire simulation models were conducted on a real wildfire in Getty, California, USA. The results show that the IGN model can accurately and explicitly describe the spatiotemporal characteristics of the wildfire spread in a novel graph form while maintaining competitive simulation refinement and computational efficiency (Jaccard: 0.587, SM: 0.740, OA: 0.800).

Under circumstances of fire, panic usually brings uncertainty and unpredictability to evacuation. Therefore, a deep understanding of panic is desired. This study aims to dig into the underlying mechanism of fire evacuation panic by measuring and analysing psycho- and physiological indicators. In the experiment, participants watched a simulated train station within which three sets of stimuli were triggered separately. Eye movement and brain haemodynamic responses were collected during the watch, while questionnaires and interviews of emotions were conducted after. The analysed physiological indicators include the amplitude of pupil dilation, the time ratios of fixation and saccade, the binned entropy of gaze location, and the brain activation coefficients. The results of this research indicate that fire evacuation panic can be broken down into two elements. (1) Unawareness of situation: less knowledge of the situation leads to a higher level of panic; (2) Intensity of visual stimulation: the panic level is escalated with increased severity of fire that is perceived.