Crowd accident surveys have shown that regardless of the initial triggering factors, pedestrian fall behavior is the most critical factor causing and aggravating crowd accidents in public traffic areas (PTAs). The application of pedestrian fall behavior detection methods in PTAs is significant. Once deployed, they would prevent many pedestrians from losing life in crowded traffic area accidents. However, most existing methods are still focused on medical assistance for the elderly. Therefore, this paper conducted bibliometric and content analyses, combining fall detection-related keywords from internationally recognized literature databases and benchmark pedestrian behavior datasets. Based on the analysis of the state-of-the-art (SOTA) achievements in fall detection methods, the fall detection methods were classified into different categories according to the research approach. This study undertakes a comprehensive analysis of five predominant methods, namely, computer vision, Internet of Things, smartphone, kinematic, and wearable device-based methods. Furthermore, the benchmark datasets, including fall scenarios, were introduced and compared. Finally, this study provides a detailed discussion of existing fall detection methods, and possible future directions are identified considering the application requirements in PTAs. This overview may help researchers understand the SOTA fall detection methods and devise new methodologies by improving and synthesizing the highlighted issues in PTAs.

Pedestrian panic behavior is a primary cause of overcrowding and stampede accidents in public micro-road network areas with high pedestrian density. However, reliably detecting such behaviors remains challenging due to their inherent complexity, variability, and stochastic nature. Current detection models often rely on single-modality features, which limits their effectiveness in complex and dynamic crowd scenarios. To overcome these limitations, this study proposes a contour-driven multimodal framework that first employs a CNN (CDNet) to estimate density maps and, by analyzing steep contour gradients, automatically delineates a candidate panic zone. Within these potential panic zones, pedestrian trajectories are analyzed through LSTM networks to capture irregular movements, such as counterflow and nonlinear wandering behaviors. Concurrently, semantic recognition based on Transformer models is utilized to identify verbal distress cues extracted through Baidu AI’s real-time speech-to-text conversion. The three embeddings are fused through a lightweight attention-enhanced MLP, enabling end-to-end inference at 40 FPS on a single GPU. To evaluate branch robustness under streaming conditions, the UCF Crowd dataset (150 videos without panic labels) is processed frame-by-frame at 25 FPS solely for density assessment, whereas full panic detection is validated on 30 real Itaewon-Stampede videos and 160 SUMO/Unity simulated emergencies that include explicit panic annotations. The proposed system achieves 91.7% accuracy and 88.2% F1 on the Itaewon set, outperforming all single- or dual-modality baselines and offering a deployable solution for proactive crowd safety monitoring in transport hubs, festivals, and other high-risk venues.

Pedestrian panic behavior is a primary cause of overcrowding and stampede accidents in public micro-road network areas with high-density pedestrians. However, reliably detecting such behaviors remains challenging due to their inherent complexity, variability, and stochastic nature. Current detection models often rely on single-modality features, limiting their effectiveness in complex, dynamic crowd scenarios. To overcome these limitations, this study proposes a novel multimodal panic detection approach integrating crowd density mapping, pedestrian trajectory analysis, and semantic recognition. Specifically, crowd density maps are generated using a convolutional neural network (CDNet) to identify regions with abnormal density gradients via contour analysis. Within these potential panic zones, pedestrian trajectories are analyzed through LSTM networks to capture irregular movements such as counterflow and nonlinear wandering behaviors. Concurrently, semantic recognition based on Transformer models is utilized to identify verbal distress cues extracted through Baidu AI real-time speech-to-text conversion. These multi-modal features—spatial, temporal, and semantic—are systematically fused and weighted using an MLP-based feature fusion framework to achieve robust panic detection accuracy. Comprehensive experiments on the UCF Crowd dataset demonstrate that this proposed approach significantly outperforms state-of-the-art methods, achieving an accuracy of 91.7%. The proposed detection framework can technically support real-time crowd safety management further in high-density pedestrian scenarios, including significant public crowd-gathering activities, transportation hubs, and emergency evacuations.