Abnormal crowd behavior recognition is one of the research hotspots in computer vision. Its goal is to use computer vision technology and abnormal behavior detection models to accurately perceive, predict, and intervene in potential abnormal behaviors of the crowd and monitor the status of the crowd system in public places in real time, to effectively prevent and deal with public security risks and ensure public life safety and social order. To this end, focusing on the abnormal crowd behavior recognition technology in the computer vision system, a systematic review study of its theory and cutting-edge technology is conducted. First, the crowd level and abnormal behaviors in public places are defined, and the challenges faced by abnormal crowd behavior recognition are expounded. Then, from the dimensions based on traditional methods and based on deep learning, the mainstream technologies of abnormal behavior recognition are discussed, and the design ideas, advantages, and limitations of various methods are analyzed. Next, the mainstream software tools are introduced to provide a comprehensive reference for the technical framework. Secondly, typical abnormal behavior datasets at home and abroad are sorted out, and the characteristics of these datasets are compared in detail from multiple perspectives such as scale, characteristics, and uses, and the performance indicators of different algorithms on the datasets are compared and analyzed. Finally, the full text is summarized and the future development direction of abnormal crowd behavior recognition technology is prospected.

In emergencies close to a T-shaped street junction (TSJ), pedestrian overcrowding and trampling are typical unstable situations triggered by external disturbances, potentially leading to serious casualties during crowd movement. It is significant to analyze the pedestrian flow stability to achieve safe and efficient crowd evacuation. To analyze the stability of crowd flow, this study uses the two-dimensional Aw-Rascle dynamic-flow model to build the pedestrian merging dynamics at a TSJ area. Then the principle of the Lyapunov stability criterion in modern control theory is employed to analyze the stability of the pedestrian crowd. A novel stability criterion for an asymmetric pedestrian merging layout at a TSJ is then proposed. When the crowd acceleration in the local area exceeds the critical acceleration, the crowd will evolve to an unstable state. To validate this criterion, numerical simulations of a case study of a crowd stampede with serious casualties (1703 victims reported) occurring at a TSJ area in Mecca, Saudi Arabia, during the Hajj in 2015 is implemented. To build stable acceleration regimes, this paper tunes the key parameters of the number of pedestrians, width ratio of main street to branch street, and pedestrian drop-off location along the street 223, and discusses the stable ranges of acceleration to satisfy the pedestrian stability criterion. This criterion and the experimental results can supply new information to control pedestrian flow and keep it stable.

Increasing maintenance costs will hinder the expansion of the wind power industry in the coming decades. Training personnel, field maintenance, and frequent boat or helicopter visits to wind turbines (WTs) is becoming a large cost. One reason for this cost is the routine turbine inspection repair and other stochastic maintenance necessitated by increasingly unbalanced figure loads and unequal turbine fatigue distribution in large-scale offshore wind farms (OWFs). In order to solve the problems of unbalanced fatigue loads and unequal turbine fatigue distribution, thereby cutting the maintenance cost, this study analyzes the disadvantages of conventional turbine fatigue definitions. We propose an improved fatigue definition that simultaneously considers the mean wind speed, wind wake turbulence, and electric power generation. Further, based on timed automata theory, a power dispatch approach is proposed to balance the fatigue loads on turbines in a wind farm. A control topology is constructed to describe the logical states of the wind farm main controller (WFMC) in an offshore wind farm. With this novel power control approach, the WFMC can re-dispatch the reference power to the wind turbines according to their cumulative fatigue value and the real wind conditions around the individual turbines in every power dispatch time interval. A workflow is also designed for the control approach implementation. Finally, to validate this proposed approach, wind data from the Horns Rev offshore wind farm in Denmark are used for a numerical simulation. All the simulation results with 3D and 2D figures illustrate that this approach is feasible to balance the loads on an offshore wind farm. Some significant implications are that this novel approach can cut the maintenance cost and also prolong the service life of OWFs.

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.

Crowd abnormal behavior recognition is one of the research hotspots in computer vision. Its goal is to use computer vision technology and abnormal behavior detection models to accurately perceive, predict, and intervene in potential abnormal behaviors of the crowd, and monitor the status of the crowd system in public places in real-time, to effectively prevent and deal with public security risks and ensure public life safety and social order. To this end, focusing on the crowd abnormal behavior recognition technology in the computer vision system, a systematic review study of its theory and cutting-edge technology is conducted. First, the crowd level and abnormal behaviors in public places are defined, and the challenges faced by crowd abnormal behavior recognition are expounded. Then, from the dimensions based on traditional methods and based on deep learning, the mainstream technologies of abnormal behavior recognition are discussed, and the design ideas, advantages, and limitations of various methods are analyzed. Next, the mainstream software tools are introduced to provide a comprehensive reference for the technical framework; Secondly, typical abnormal behavior datasets at home and abroad are sorted out, and the characteristics of these datasets are compared in detail from multiple perspectives such as scale, characteristics and uses, and the performance indicators of different algorithms on the datasets are compared and analyzed; Finally, the full text is summarized and the future development direction of crowd abnormal behavior recognition technology is prospected.

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.