This paper presents a crowd-density simulation system developed using the Unity engine to predict and prevent crowd-crush accidents. Crowd-crush accidents can occur in various environments where large crowds gather, often leading to significant casualties. Effective tools for risk assessment, training, and decision-making are crucial in preventing such accidents. The proposed simulator aims to provide a realistic virtual environment that accurately models crowd behavior and dynamics. By analyzing crowd density, flow patterns, and bottleneck situations, the simulator enables users to identify potential hazards, evaluate evacuation strategies, and develop proactive crowd management measures. The system incorporates advanced features such as real-time crowd visualization, interactive scenario creation, and data analytics. Validation experiments demonstrate the effectiveness of the simulator in predicting crowd-crush risks and supporting decision-making processes. The proposed solution enhances crowd safety and helps prevent accidents in various public-gathering scenarios.

This paper set out to revise and improve existing autonomous driving models using reinforcement learning, thus proposing a reinforced autonomous driving prediction model. The paper conducted training for a reinforcement learning model using DQN, a reinforcement learning algorithm. The main aim of this paper was to reduce the time spent on training and improve self-driving performance. Rewards for reinforcement learning agents were developed to mimic human driving behavior as much as possible. High rewards were given for greater distance travelled within lanes and higher speed. Negative rewards were given when a vehicle crossed into other lanes or had a collision. Performance evaluation was carried out in urban environments without pedestrians. The performance test results show that the model with the collision prevention model exhibited faster performance improvement within the same time compared to when the model was not applied. However, vulnerabilities to factors such as pedestrians and vehicles approaching from the side were not addressed, and the lack of stability in the definition of compensation functions and limitations with respect to the excessive use of memory were shown.

This study, a model is proposed that generates descriptive sentences for input images visually impaired individuals. For this purpose, a novel image captioning approach is introduced, integrating the principles of human visual Understanding mechanisms with the Vision Transformer (ViT) architecture, further enhanced by deep reinforcement learning. First, features are extracted from the image based on human visual perception. Second, the image features are encoded through the encoding block of ViT and input into the long short-term memory (LSTM) network to generate annotations for the image. Finally, reinforcement learning is optimized to further enhance the accuracy of the generated captions. Evaluations were performed utilizing the MSR-VTT benchmark dataset, which is widely used for image captioning tasks. Experimental results on the MSR-VTT benchmark dataset demonstrate that the proposed model achieves BLEU-4 of 43.0, METEOR of 29.1, ROUGE-L of 62.7, and CIDEr-D of 54.9, surpassing state-of-the-art baseline models across all evaluation metrics. The proposed model can be applied to video annotation applications for the visually impaired. In contrast to prior works that primarily rely on conventional convolutional architectures, the proposed model uniquely incorporates human-inspired visual perception principles and Vision Transformer-based global encoding, offering a novel and interpretable framework tailored for assistive image captioning.