In the field of computer vision, convolutional neural network (CNN)-based models have demonstrated high accuracy and good generalization performance. However, in semantic segmentation, CNN-based models have a problem—the spatial and global context information is lost owing to a decrease in resolution during feature extraction. High-resolution networks (HRNets) can resolve this problem by keeping high-resolution processing layers parallel. However, information loss still occurs. Therefore, in this study, we propose an HRNet combined with an attention module to address the issue of information loss. The attention module is strategically placed immediately after each convolution to alleviate information loss by emphasizing the information retained at each stage. To achieve this, we employed a squeeze-and-excitation (SE) block as the attention module, which can seamlessly integrate into any model and enhance the performance without imposing significant parameter increases. It emphasizes the spatial and global context information by compressing and recalibrating features through global average pooling (GAP). A performance comparison between the existing HRNet model and the proposed model using various datasets show that the mean class-wise intersection over union (mIoU) and mean pixel accuracy (MeanACC) improved with the proposed model, however, there was a small increase in the number of parameters. With cityscapes dataset, MeanACC decreased by 0.1% with the proposed model compared to the baseline model, but mIoU increased by 0.5%. With the LIP dataset, the MeanACC and mIoU increased by 0.3% and 0.4%, respectively. The mIoU also decreased by 0.1% with the PASCAL Context dataset, whereas the MeanACC increased by 0.7%. Overall, the proposed model showed improved performance compared to the existing model.

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.

Semantic segmentation is a fundamental task in computer vision. Over time, efforts have been made to mitigate the loss of spatial and global context information due to resolution reduction, starting with Fully Convolutional Networks (FCN). However, Convolutional Neural Networks (CNN) models still face challenges, as the down-sampling process reduces resolution, leading to the loss of essential spatial and global context information. In this paper, we combine the HRNet model, known for its exceptional performance in semantic segmentation, with the Channel and Spatial Attention (CSA) module. The CSA module highlights important spatial and global context information in semantic segmentation. It is structured so that the result of channel attention becomes the input for spatial attention. Channel attention recalibrates each channel's resolution into a scalar value using a Squeeze-and-Excitation (SE) block, which is then multiplied by the original feature map. Spatial attention employs the Polarized Self-attention (PSA) module to compress the feature map channels from channel attention into a single resolution, and recalibrates them by multiplying by the input feature map. To assess the semantic segmentation performance of the proposed model, we compare and evaluate its performance with or without the CSA module, and against various models, using the LIP and PASCAL Context datasets. As a result of the experiment, in the LIP data set, mean Intersection over Union (mIoU) was improved by 0.5%, and Mean Pixel Accuracy (MPA) by 0.1%, improving both segmentation and class classification accuracy. In the PASCAL Context 59 data set, mIoU was maintained, while MPA was improved by 0.2%, which improved class classification accuracy. In the PASCAL Context 60 data set, mIoU decreased by 0.1%, while MPA was improved by 0.2%, improving class classification accuracy.