Search

In order to quickly and efficiently simulate the crowd movement process under the evacuation scene, so as to reduce the casualties in emergency evacuation, this article proposes an indoor emergency evacuation model based on the improved artificial swarm algorithm. In this article, the cellular automata (CA) model is used to establish the evacuation environment, and then the artificial bee colony (ABC) algorithm is used to simulate the crowd evacuation. However, due to too many obstacles existing in the neighboring cells of some position, the individual in that position have to wait. If the distance equation is used to calculate the fitness, it may cause other individuals to choose these positions as new position and repeat the same mistake above. To reduce the occurrence of such cases, we improve the fitness function of the ABC algorithm. In the fitness function, the factors of attraction and repulsion force in social force model are introduced. And on the basis of the ABC algorithm, we propose the visual employed bee. The visual employed bee leads the onlooker bee to evacuate, so as to improve the efficiency of evacuation. The research results of this article can provide ideas for evacuation modeling and useful guidance for formulating evacuation strategies to reduce evacuation time and disaster losses.

Abnormal event detection, as a hot research field in intelligent video monitoring system, has attracted many researchers' attention in recent years. In order to overcome the shortcomings of the semi-supervised model, namely the training sample is difficult to contain all possible situations, leading to the occurrence of error detection, we propose a method based on support vector data description (SVDD). The principle of the method is to train the model with normal data and abnormal data respectively to obtain two SVDD models, and then judge whether there are abnormal events according to the results of the two models. This method has been tested by existing data sets and achieved good results.

Emergent events and catastrophe accidents in large public places are increasing in recent years. In order to ensure successful escape from the accident places and provide a scientific and efficient plan for evacuees, an evacuation model based on artificial fish swarm algorithm(AFSA) for large public place is presented in this paper. In the model, each evacuee is considered as an intelligent artificial fish, and the foraging process of fish swarm is used to simulate evacuation behavior. Route choice strategy is proposed to simulate the psychological behaviors of evacuation individuals. Evacuation Simulation algorithm based on artificial fish swarm is proposed and applied to large stadium evacuation problem. The experimental and simulated results show that the presented model and algorithm can simulate the characteristics of the evacuation individuals, the evacuation rules and the uncertain factors in the evacuation process. The modelling approach has good effect in evacuation problem.