Situational Awareness Research Articles

Utilizing Robotic System for Disaster Scene Classification - A Deep Learning Approach

Disaster scene classification plays a vital role in emergency management by facilitating rapid assessment and response to scenarios such as floods, earthquakes, and wildfires. Traditional image classification methods face challenges due to the complexity and variability of disaster scenes, which often include irregular patterns and diverse environmental factors. In recent years, deep learning, particularly convolutional neural networks (CNNs), has demonstrated significant potential in improving the accuracy of disaster scene classification. This project integrates a CNN-based approach with a remote-controlled robot equipped with real-time image-capturing capabilities. The robot navigates disaster zones, capturing images that are processed using CNN architecture like VGG16 and VGG19 to classify disaster scenes efficiently. The robotic system enhances situational awareness by autonomously collecting vital information in hazardous environments, transmitting real-time data for classification, and providing timely insights for emergency response. This integration of robotics with deep learning not only automates disaster scene classification but also reduces reliance on large, labeled datasets, improving performance and response effectiveness.

Crowd Management in Railway System Using Deep Learning

As urbanization progresses and greater demands for public security are imposed by busy surroundings like train stations, conventional surveillance systems become handicapped by core shortcomings in terms of responsiveness, scalability, and effectiveness. Most of these conventional systems are human-operated for real-time surveillance of CCTV live feeds, and it is prone to lag in response, to exhaustion, as well as the oversight of abnormalities owing to a voluminous stream of data. This project overcomes these challenges by taking advantage of Artificial Intelligence (AI) and Machine Learning (ML) to optimize real-time crowd management through the YOLOv5 object detection algorithm. The primary aim is to transform passive surveillance networks into proactive monitoring systems that can automatically detect crowd density, detect abnormal movement patterns, and trigger timely alerts without human intervention. YOLOv5 is chosen due to its high detection rates, speed, and light footprint, making it ideal for real-time video inspection on edge or low-power systems. The system inspects real-time video feeds from current CCTV equipment, identifies and tracks persons, and generates alerts upon threshold breaches of crowd levels or established behavioral abnormalities. The implementation of this AI model on existing surveillance arrangements is a valuable addition to the area of operational safety, being a cost-optimized, scale-up, technology-enabled solution towards crime prevention and crowd management. Not only is the system improved situational awareness but also eased the load of human resources with increased decision support through actionable insight. This study substantiates the efficiency of the designed solution via simulation and live trials, with significantly improved response times for threat detection, surveillance coverage, and general security performance. In addition, the modularity of the system makes it convenient to scale and adapt across environments such as transportation centers, arenas, and crowded public areas

Krishi Prahari: An IoT and ML-Based Framework for Smart Agricultural Threat Detection and Resource Optimization

Agriculture remains the backbone of India’s economy, but it is increasingly vulnerable to a variety of critical threats such as climate variability, crop fires, wild animal intrusions, and poor resource utilization. These challenges not only reduce crop yield but also lead to long-term damage to soil fertility, farmer income instability, and increased water consumption. In this context, the paper introduces Krishi Prahari, an intelligent, IoT and Machine Learning-powered agricultural management system developed to detect threats at an early stage and optimize decision-making processes in real time. Using an array of linked sensors, which input data into machine learning algorithms for predictive analysis, the system is meant to constantly monitor environmental conditions. This provides for early alerts for fire threats, tracking of animal movement in agricultural areas, and irrigation optimization dependant on soil conditions. By providing a proactive, automated, low-cost strategy adaptable across multiple climatic and geographical zones, the recommended solution fills in the deficiencies in existing reactive solutions. Water consumption efficiency has increased considerably, crop protection from animals and fire has been boosted, and local farmers' simplicity of use has been proven by rigorous field testing. By strengthening farm-level situational awareness and resource management capacities, Krishi Prahari consequently provides a crucial step toward obtaining climate-resilient and precision agriculture in India [1].

Military Vehicle Object Detection System

Abstract: The detection of military vehicles is critical for modern defense systems and surveillance technologies. In this project, we propose a deep learning-based object detection system capable of identifying and classifying military vehicles under realworld constraints such as varied terrain, lighting, and camouflage. Our method incorporates hierarchical feature representation and post-processing strategies including non-maximum suppression. A custom military vehicle dataset (MVD) wasdeveloped fortraining and testing. The proposed system achieves robust real-time detection and can be deployed across a variety of defense applications. The Military Vehicle Object Detection project aims to enhance the situational awareness of military operations by developinganautomatedsystemcapableof detecting, classifying, and tracking military vehicles in real-time. Leveraging advanced deep learning techniques, such asconvolutionalneuralnetworks(CNNs), the system processes live video feeds or static images to accurately identify various types of military vehicles, including tanks and armored personnel carriers. The implementation includes a user-friendly web interface for operators to visualize detection results and access historical data. Additionally, a number platedetectionmoduleenhances security by extracting and verifying vehicle identification information. This project contributes to improved operational efficiencyanddecision-makinginmilitary contexts by facilitating rapid identification and monitoring of military assets

Design and Implementation of a Remotely Operated Vehicle for Real-Time Monitoring Application

Efficient surveillance of large-scale commercial warehouses and private properties guarantees asset security, operational lapse, and real-time situational awareness. Conventional surveillance systems often lack mobility and flexibility, which reduces their effectiveness in dynamic environments. This paper discusses the design and implementation of a flexible, low-cost, and energy-efficient Remotely Operated Vehicle (ROV) developed and designed for real-time remote surveillance of spacious commercial premises. The main goal of this project is to facilitate remote monitoring with real-time video streaming, providing an affordable option compared to fixed camera setups and costly autonomous patrol robots. The suggested system uses an STM32F4xx series microcontroller for real-time management, a 2.4 GHz NRF24L01 transceiver for wireless connectivity, and a Python-driven video streaming module combined with an HTML-based control interface reachable over the local network. The power system incorporates a 76 Wh LiFePO₄ battery, with stable voltage regulation achieved using LM2596 and AMS1117 modules for different subsystems. The software architecture combines Arduino C, Python, and HTML, allowing effective control, live feed management, and user-friendly access. The system’s novelty lies in its hybrid multi-platform design, wireless range, and energy efficiency, tailored for indoor and semi-outdoor commercial environments. Experimental results show that the ROV achieves a stable control range of over 50 meters and continuous operation exceeding 1.5 hours on a single charge. Video streaming performance remains consistent with minimal possible latency, even in warehouse environments with moderate interference. While currently designed for commercial and industrial use, the system’s modularity allows future integration of features such as autonomous navigation, object detection, and deployment in hazardous and defence-sensitive environments. In conclusion, the proposed ROV offers a scalable and practical solution for commercial surveillance, contributing towards intelligent, mobile, and energy-efficient monitoring systems.

Visual Behavior Analysis

Abstract: With rising crime rates and increasing security concerns, intelligent surveillance has become crucial for ensuring public safety.This paper presents a Visual Behavior Analysis system that automates crime detection and tracking using real-time video surveillance. The system employs You Only Look Once (YOLO) is a deep learning algorithm recognized for its speed and accuracy in multi-object detection, to monitor public spaces efficiently. By analysing movement patterns and identifying anomalies, it can detect potential threats and generate instant alerts to prevent criminal activities. Additionally, the system integrates facial recognition through the use of Haar Cascade classifiers to assist law enforcement in identifying individuals involved in suspicious activities. It also detects objects such as weapons or unusual postures, improving the accuracy of crime prediction. The system processes live video feeds, minimizing response time and reducing reliance on manual monitoring. By leveraging computer vision and artificial intelligence, improving security measures by offering proactive crime prevention and real-time situational awareness. It goes beyond just aids in immediate threat detection but also supports post-incident investigations by preserving crucial evidence. The proposed system provides a scalable, efficient, and automated solution for modern surveillance challenges, contributing to a safer and smarter security infrastructure.

Situational Awareness in Leadership: Application of Methods in Business Organisations

This study analysed the leadership approaches that determine one of the components of organisational resilience: situational awareness in business organisations. A lack of situational awareness in leadership results in poor decision making and low organisational resilience, which undermines the continuity and sustainability of the organisation’s activities. This observation prompts the following research question: which leadership methods enhance situational awareness and how are these methods effectively applied in business organisations? This study analysed the situational awareness requirements for leadership through leadership methods. With the help of mixed methods that integrate qualitative and quantitative approaches, an empirical study was conducted in eight European countries; in total, 30 leaders of business organisations were interviewed and 3370 employee questionnaires were analysed. The analysis identified the leadership methods that enhance situational awareness; it also presented the assumptions that determine the effectiveness of these methods. The relationship between leadership methods and situational awareness was found to be mediated by the interaction of the two elements of situational awareness with twelve leadership methods. These findings provide a structured approach to explaining how leadership methods affect situational awareness, thus complementing existing theoretical frameworks and encouraging the development of new theoretical models.

Alan Turing and Gordon Welchman’s ultra intelligence in the new space age of ultra

Abstract Ultra intelligence is a complex multifaceted process that consumes enormous amounts of manpower and coordination 24/7 and it has its roots in the work of two intelligence pioneers, Alan Turing and Gordon Welchman, that dates to the World War II and before that. Back then success relied heavily on the wits of these two pioneers and Agnes, the electromechanical machine they developed. Back then a two-hour response was a norm. Modern ultra intelligence is expected to deliver quick decision advantages at machine speed which in turn makes operations effective. To achieve this, modern ultra intelligence uses several key technologies: Artificial Intelligence and Machine Learning for quickly identifying trends, anomalies, and risks; Command and Control for establishing quick flows of accurate and complete information for real-time decision-making, situational awareness, operational planning, resource allocation optimization, and improved communications; Data Security and Integration for protecting against cyber threats and data breaches and keeping sensitive information and critical infrastructures secure; Data Analytics and Fusion for making real-time data-driven decisions for improving performance, maximizing efficiencies, and minimizing risk; and Diverse Communication Platforms. How did ultra intelligence advance from such humble beginnings to the space age of ultra? This paper considers.

Assessment of Flight Operations among IAU Aviation Students at Mactan and Ormoc Aerodromes

This study assessed and compared the flight operations of Indiana Aerospace University (IAU) aviation students designated at Mactan and Ormoc aerodromes for the academic year 2023–2024. Utilizing a descriptive quantitative research design, the study evaluated three primary areas: flight training performance, performance metrics, and airport operations. A self-made questionnaire using a 5-point Likert scale was distributed to 50 respondents, composed of 44 aviation students and 6 flight instructors with flight experience at both aerodromes. Additionally, qualitative insights were gathered from five participants through structured interviews to supplement the statistical findings. Results revealed that while both aerodromes contributed positively to the flight training experience, Mactan Aerodrome was generally perceived as the more favorable training environment. Students cited Mactan’s controlled airport status, real-time air traffic communication, and higher traffic density as key factors in improving situational awareness, decision-making skills, and operational proficiency. These real-world conditions mirrored actual aviation industry demands, giving students practical exposure to complex flight environments. In contrast, Ormoc Aerodrome was recognized for its contribution to foundational flying skills such as takeoff, landing, and airwork but offered limited exposure to complex scenarios due to its uncontrolled, low-traffic nature. The study concluded that aerodrome conditions significantly impact pilot training effectiveness and recommended targeted improvements. These included extending Ormoc’s operational hours, introducing controlled flight scenarios, upgrading infrastructure, and enhancing simulation exercises. For Mactan, optimizing flight planning sessions, reinforcing technical skill development, and integrating fuel efficiency strategies were advised. These findings provide valuable input for enhancing IAU’s training program and ensuring students are equipped with the competencies required for modern aviation practice.

AI AS A CO-PILOT (Enhancing Aviation Safety and Efficiency while Keeping Human Pilots at the Center)

ABSTRACT This study investigates the integration of Artificial Intelligence (AI) as a co-pilot in aviation. As aviation increasingly embraces automation, concerns about safety, pilot roles, and ethical implications grow. The research emphasizes the necessity for human-AI collaboration in which AI enhances pilot capabilities without replacing them. Using a mixed-methods approach that includes literature review, case analysis, and stakeholder input, the study provides a model for safely integrating AI into flight operations. Key findings support that AI should assist rather than replace pilots, maintaining human oversight in high-stakes environments. The thesis concludes with policy and design recommendations for AI integration in aviation. This comprehensive study explores the transformative potential of Artificial Intelligence (AI) in modern commercial aviation, specifically emphasizing the concept of AI as a co-pilot. With rising operational complexity, global pilot shortages, and increasing demand for enhanced safety and efficiency, AI presents itself as a promising technological innovation. It has the capacity to provide real-time flight data interpretation, system anomaly alerts, autonomous recommendations, and task automation that collectively enhance situational awareness and reduce pilot workload. However, the expansion of AI usage in aviation also introduces critical concerns around pilot-AI interaction, explainability of AI systems, cybersecurity risks, and ethical responsibility in cases of failure. The research argues for a balanced integration model where AI supports—rather than supplants—human pilots. The idea of a "co-pilot AI" is conceptualized to highlight systems that work in tandem with human decision-makers, offering real-time insights while respecting human authority. Human-in-the-loop (HITL) design frameworks, explainable AI (XAI) principles, and trust-calibration models were central to the theoretical foundation of this research. A mixed-methods approach was adopted to strengthen the analysis, combining literature reviews, international aviation policy reviews, expert interviews, and comparative incident analysis. The findings of the study reveal a general consensus among professionals that while AI significantly enhances operational responsiveness and data processing capabilities, human oversight remains non-negotiable for moral judgment and adaptive decision-making during complex flight conditions. The research further identifies that system transparency, pilot familiarity, and regulatory clarity are essential for cultivating pilot trust in AI tools. It also highlights training gaps in existing aviation programs, where pilots are inadequately prepared to interface effectively with AI technologies. From a managerial and policy standpoint, the study recommends a strategic shift in training curricula to include AI-system literacy, greater investment in explainable interfaces, cross-industry collaboration for best practices, and harmonization of international regulatory standards. These measures, if implemented, will not only prevent technological overdependence but also build a more resilient and adaptive aviation ecosystem.

A systematic review on human-AI hybrid systems and human factors in air traffic management

Artificial intelligence (AI) is poised to play a transformative role in supporting human air traffic controllers, enabling them to manage increasing traffic amidst growing capacity pressures. By enhancing their ability to manage increasing air traffic demands under growing capacity pressures, AI holds the potential to transform this safety-critical and human-dependent domain. To explore this potential, this systematic review compiles and synthesises studies on human-AI interactions in air traffic management (ATM), aiming to (1) examine the characteristics of human-AI hybrid (HAH) systems, (2) identify relevant human factors studies and their contributions, and (3) derive guidelines for future advancements and experimental designs. Of the 125 studies that met the inclusion criteria, two primary themes were identified: HAH systems and human factors. This review explores the use of HAH systems in ATM as enhancement tools across different operational levels, detailing their implementation through stages of conceptualisation, development, evaluation, and training. Additionally, it examines key human factor dimensions – such as workload, situation awareness, vigilance, decision-making, teamwork, communication, acceptance, and trust – by presenting relevant measurements, findings, and their implications for improving HAH and human-AI collaboration efficiency in ATM systems. This review offers suggestions for future advancements in HAH systems, ultimately contributing to safer and more efficient ATM.

Artificial Intelligence and Coronary Artery Bypass Grafting: Current Status and Future Perspectives

Artificial intelligence (AI) is revolutionizing the field of coronary artery bypass grafting (CABG) by enhancing various stages of the surgical process. Pre-operatively, AI, particularly through machine learning (ML) and language processing (LP), assists in consultations, medical diagnostics, and clinical predictions. ML models analyze patient data to predict outcomes and stratify risks, while LP automates the documentation of patient interactions, improving efficiency and reducing recall bias. Intra-operatively, computer vision (CV) plays a crucial role in improving surgical performance and team dynamics. CV can automate surgical checklists, assist surgeons by providing real-time feedback, and enhance procedural accuracy. It also aids in instrument tracking and situational awareness, contributing to better team coordination and reduced intraoperative errors. These applications are particularly beneficial for surgical trainees, offering guidance and improving their technique through real-time analysis. Post-operatively, AI continues to support patient care by predicting complications and optimizing recovery plans. ML models assess the risk of post-operative complications, such as major bleeding, myocardial infarction, and acute kidney injury, based on pre-operative characteristics. This enables personalized patient management and targeted interventions to mitigate risks. Additionally, CV can streamline post-operative processes by monitoring patient turnover and improving operating room efficiency. Despite its potential, the integration of AI in CABG faces challenges, including model overfitting, lack of transparency, and high implementation costs. Ethical considerations, such as patient privacy and data security, must be addressed to ensure responsible AI use. Future research should focus on validating AI applications in real-world settings and exploring their impact on minimally invasive techniques and overall surgical outcomes.

AI-Driven Threat Intelligence for Predicting Advanced Persistent Attacks in Cloud-Based IT Services

The adoption of cloud-based IT services has transformed modern enterprise operations, offering flexibility and scalability. However, this evolution has also introduced significant security challenges, particularly from Advanced Persistent Threats (APTs), which are sophisticated, stealthy, and often long-lasting attacks designed to bypass conventional defence mechanisms. Addressing such threats requires a forward-looking approach that emphasizes prediction and early intervention rather than reactive countermeasures. This research presents an innovative artificial intelligence (AI)-based framework that combines threat intelligence with deep learning models to anticipate and detect APTs in cloud environments. The proposed system employs Long Short-Term Memory Autoencoders (LSTM-AE) to uncover abnormal patterns in system behaviours by analysing multiple data sources, including network traffic, system logs, and threat intelligence feeds. The framework is trained and evaluated using publicly available datasets such as CICIDS 2017, along with custom cloud log data. The results highlight the model's ability to achieve high detection accuracy while minimizing false positive rates, outperforming traditional intrusion detection approaches. By integrating contextual threat intelligence with AI-based behavioural analysis, the framework enhances real-time situational awareness and supports proactive cybersecurity measures. This study contributes a scalable and adaptive solution for strengthening cloud infrastructure against evolving and complex threat scenarios.

The impact of signal variability on COVID-19 epidemic growth rate estimation from wastewater surveillance data.

Testing samples of wastewater for markers of infectious disease became a widespread method of surveillance during the COVID-19 pandemic. While these data generally correlate well with other indicators of national prevalence, samples that cover localised regions tend to be highly variable over short time scales. Here we introduce a procedure for estimating the real-time growth rate of pathogen prevalence using time series data from wastewater sampling. The number of copies of a target gene found in a sample is modelled as time-dependent random variable whose distribution is estimated using maximum likelihood. The output depends on a hyperparameter that controls the sensitivity to variability in the underlying data. We apply this procedure to data reporting the number of copies of the N1 gene of SARS-CoV-2 collected at water treatment works across Scotland between February 2021 and February 2023. The real-time growth rate of the SARS-CoV-2 prevalence is estimated at all 121 wastewater sampling sites covering a diverse range of locations and population sizes. We find that the sensitivity of the fitting procedure to natural variability determines its reliability in detecting the early stages of an epidemic wave. Applying the same procedure to hospital admissions data, we find that changes in the growth rate are detected an average of 2 days earlier in wastewater than in hospital admissions. In conclusion, this paper provides a robust method to generate reliable estimates of epidemic growth from highly variable data. Applying this method to samples collected at wastewater treatment works provides highly responsive situational awareness to inform public health.

A Multiscale Approach to Cyber-Mechanical Threat Modeling for Predicting and Preventing Failures in Critical Energy Infrastructure

The increasing convergence of cyber and mechanical domains in critical energy infrastructure, including electric power grids, oil and gas pipelines, and renewable energy systems, has significantly expanded the attack surface for sophisticated cyber threats, particularly ransomware. These hybrid systems, known as cyber-mechanical systems (CMS), present complex vulnerabilities across multiple temporal and spatial scales, which are inadequately addressed by traditional security frameworks that treat cyber and physical layers in isolation. This study introduces a novel multiscale AI/ML-based framework for predictive resilience modeling and real-time anomaly detection in CMS environments. The proposed architecture integrates a Convolutional Long Short-Term Memory 3D (ConvLSTM3D) model for high-resolution spatiotemporal anomaly detection, and a Graph Neural Network (GNN) for dynamic threat propagation analysis across system hierarchies. The framework was evaluated using both synthetic simulations and the CICIDS 2017 dataset, yielding a test accuracy of 88.86%, an AUC of 0.89, and an F1-score of 0.38 in highly imbalanced ransomware detection scenarios. A simulated ransomware attack on a SCADA-controlled energy network demonstrated the model’s ability to detect threats at the micro (≤1s), meso (1s–1h), and macro (>1h) levels, with detection precision exceeding 95% for short-duration anomalies. These results confirm that modeling cyber-mechanical interactions across multiple scales significantly enhances early threat detection and supports situational awareness. Future research should explore federated learning, continual adaptation, and explainable AI to enable real-time deployment and broader generalizability. By bridging cyber-physical modeling, machine learning, and resilience engineering, this study contributes an actionable framework for safeguarding critical energy infrastructure from increasingly sophisticated and coordinated cyber threats.

FURIOUS: Fully unified risk-assessment with interactive operational user system for vessels.

Ship collision risk assessment has advanced over recent years, enhancing maritime safety. However, existing studies often describe ship domains and collision risk assessments in a static manner, lacking interactivity. Interactive visualization of collision risk, especially in multi-ship scenarios has not been sufficiently developed. This gap prompted the development of "FURIOUS: Fully Unified Risk-assessment with Interactive Operational User System for vessels." This tool aids in visualizing and analyzing collision risk of multi-ship encounter situation through real-time visualization. Our system processes data from Automatic Identification System (AIS). The system performs ship domain calculations and collision risk assessments supported by geographical computations, and includes features like real-time vessel display and collision type detection. Interactive and user-selectable elements, along with dynamic maps enhance real-time decision-making to ensure navigation safety. Additionally, the system aids both experienced and novice users in understanding complicated maritime dynamic environments. Users can adjust parameters like ship type, ship IDs, time window and map type for tailored analyses and proactive collision avoidance. We conducted a user study to validate these features, confirming that they effectively improve situational awareness and enhance decision-making capabilities in real-world scenarios. This paper details the design, implementation, and evaluation of this tool, highlighting its potential to transform maritime decision-making by improving situational awareness and enhancing operational efficiency.

Synergies in the Skies: Situation Awareness and Shared Mental Model in Digital-Human Air Traffic Control Teams

With increasing air traffic, the workload of air traffic controllers (ATCOs) and their limited number is again a restricting factor for the evolution of airspace management. Currently, possibilities to apply artificial intelligence for improving the support of ATCOs are widely discussed. By introducing a digital ATCO as a team partner for a human ATCO, we can expand capabilities. It can be trained to manage traffic across various airspace sectors without the limitations imposed by required licenses. This way, shortages of human ATCOs may be absorbed, and flexible assignment to sectors is facilitated with a digital ATCO partner. To be effective, the digital ATCO needs an understanding of current and future traffic situations to share the situation awareness of the human ATCO. The goal is to equip the digital ATCO with a comparable understanding—referred as a “mental model”—of the traffic situation and human actions, thereby improving decision-making and build up adequate trust with humans. In this work, decisive factors of traffic and management for the creation of digital situation awareness are identified and examined for their relevance and applicability for digital ATCOs. Within this study, a data-driven process of building up digital situation awareness including the influencing factors are suggested, and the usability of factors like the airspace complexity for indicating digital situation awareness are proposed. Finally, an example is presented and discussed to showcase our approach with focus on the integration of digital and human ATCOs through shared situation awareness.

Manual Expression of Breastmilk: Performing, Teaching, and Modifications of Marmet Technique

This is the second article of a three-article series that grew from the initial invitation by a member of the Clinical Lactation editorial staff to accompany a research article they had received that studied the outcomes of milk supply of mothers of preemies who implemented Marmet Technique. This research determined that the milk supply statistically significantly increased after using Marmet Technique for only one to three days (Kotsu et al., 2025). While the first article offered details on the history and importance of Marmet Technique, this second article focuses on when to use, how to perform, how to teach, and who should know the Marmet Technique, as well as modified Marmet Technique, and the use of ice and heat for engorgement and mastitis. This article lends itself to the development of situational awareness for Marmet Technique application and the technical skills necessary to perform it effectively to increase, maintain, or decrease a milk supply. Sometimes the best way to confirm competency is to teach but to teach well requires planning and a competency of its own. While teaching a large group might be rare for a lactation consultant, being able to teach this skill to an individual patient should be required. The third and final article of this series follows with additional information on integrating tools for expression and feeding expressed milk back to an infant.

Foreign experience of using video surveillance systems by law enforcement agencies in combating crime

In today’s world, video surveillance is a key element of law enforcement, that helps prevent and solve crimes, detect offenses and increase the level of security of citizens. Combined with new video analytics technologies (face and object recognition, motion detection, people counting, unauthorized access detection), surveillance cameras allow you to identify, control, and track people and vehicles. That is why they have become widely used by law enforcement officers to monitor public places, prevent crimes, monitor compliance with traffic rules, solve crimes, and identify criminals. The author analyzed the best foreign experience of implementing modern information technologies by law enforcement agencies to combat crime. In particular, the author examines the use of video surveillance by Chinese law enforcement agencies, which is used by the police to control public safety, search for suspects, monitor public events, and even implement a social rating system. The article also highlights the use of video surveillance systems for the purpose of investigating terrorist activities in the United Kingdom, namely, in identifying and charging persons involved in criminal activities. The main capabilities of the Singapore Police Force (SPF) police camera network (PolCam) are considered, which not only help prevent and solve crimes, but also expand the capabilities of the police in terms of situational awareness and allow officers to respond more quickly to incidents related to violations of law and order on the ground. The latest technological capabilities of the Israeli police in the field of control over the movement of citizens and in the field of road safety are analyzed in detail. The specifics of the use of MONOcam «smart» radars by the German police, which are used to record many traffic offenses, are highlighted. Based on the analysis, the author proposes to introduce modern video surveillance technologies into the activities of the National Police of Ukraine.

AI and IoT-Based Frameworks for Real-Time Crowd Monitoring and Security

Effective crowd management is critical for ensuring public safety during large-scale events and in densely populated urban environments. Recent advances in deep learning and computer vision have enabled real-time crowd behavior analysis, including the detection of abnormal actions such as pushing, which can lead to dangerous situations. This paper presents a review of cloud-based deep learning frameworks, focusing on the use of convolutional neural networks (CNN) and optical flow models for early detection of pushing behavior in crowded event entrances. We discuss the integration of pre-trained deep models with live video stream processing to achieve high accuracy and low latency. Existing datasets and evaluation metrics are examined, with reported detection accuracies reaching up to 87%. The review also highlights challenges such as data privacy, real-time processing constraints, and the need for comprehensive models that consider multiple behavioral and environmental factors. Finally, future directions are proposed for developing autonomous crowd safety systems that mimic human situational awareness in complex urban settings.