Search Results (115)

Blind and visually impaired (BVI) people face significant challenges in perception, navigation, and safety during travel. Existing infrastructure (e.g., blind lanes) and traditional aids (e.g., walking sticks, basic audio feedback) provide limited flexibility and interactivity for complex environments. To solve this problem, we propose a real-time travel assistance system based on deep learning. The hardware comprises an NVIDIA Jetson Nano controller, an Intel D435i depth camera for environmental sensing, and SG90 servo motors for feedback. To address embedded device computational constraints, we developed a lightweight object detection and segmentation algorithm. Key innovations include a multi-scale attention feature extraction backbone, a dual-stream fusion module incorporating the Mamba architecture, and adaptive context-aware detection/segmentation heads. This design ensures high computational efficiency and real-time performance. The system workflow is as follows: (1) the D435i captures real-time environmental data; (2) the processor analyzes this data, converting obstacle distances and path deviations into electrical signals; (3) servo motors deliver vibratory feedback for guidance and alerts. Preliminary tests confirm that the system can effectively detect obstacles and correct path deviations in real time, suggesting its potential to assist BVI users. However, as this is a work in progress, comprehensive field trials with BVI participants are required to fully validate its efficacy. Full article

As urban air mobility moves rapidly toward real-world deployment, accurate vehicle trajectory prediction and early collision risk detection are vital for safe low-altitude operations. This study presents a deep learning framework based on an LSTM–Attention network that captures both short-term flight dynamics and long-range dependencies in trajectory data. The model is trained on fifty-six routes generated from a UAM planned commercialization network, sampled at 0.1 s intervals. To unify spatial dimensions, the model uses Earth-Centered Earth-Fixed (ECEF) coordinates, enabling efficient Euclidean distance calculations. The trajectory prediction component achieves an RMSE of 0.2172, MAE of 0.1668, and MSE of 0.0524. The collision classification module built on the LSTM–Attention prediction backbone delivers an accuracy of 0.9881. Analysis of attention weight distributions reveals which temporal segments most influence model outputs, enhancing interpretability and guiding future refinements. Moreover, this model is embedded within the Short-Term Conflict Alert component of the Safety Nets module in the UAM traffic management system to provide continuous trajectory prediction and collision risk assessment, supporting proactive traffic control. The system exhibits robust generalizability on unseen scenarios and offers a scalable foundation for enhancing operational safety. Validation currently excludes environmental disturbances such as wind, physical obstacles, and real-world flight logs. Future work will incorporate atmospheric variability, sensor and communication uncertainties, and obstacle detection inputs to advance toward a fully integrated traffic management solution with comprehensive situational awareness. Full article

Historical districts face pressing disaster preparedness challenges due to their special spatial properties—risks compounded by static approaches that overlook dynamic fire–pedestrian interactions. This study employs an agent-based model (ABM) for fire simulations and AnyLogic pedestrian dynamics to address these gaps in Dukezong Ancient Town, Yunnan Province, China, considering diverse ignition points, seasonal temperatures, and wind conditions. Dynamic simulations of 16 scenarios reveal critical spatial impacts: within 30 min, ≥28% of streets became impassable, with central ignition points causing faster obstructions. Static models underestimate evacuation durations by up to 135%, neglecting early stage congestions and detours caused by high-temperature zones. Congestions are concentrated along main east–west arterial roads, worsening with longer warning distances. A mismatch between evacuation flows and shelter capacity is found. Thus, a three-stage interaction simplification is derived: localized detours (0–10 min), congestion-driven delays on critical roads (11–30 min), and prolonged structural damage afterward. This study challenges static approaches by highlighting the “fast alert-fast congestion” paradox, where rapid alerts overwhelm narrow pathways. Solutions prioritize multi-route guidance systems, optimized shelter access points, and real-time information dissemination to reduce bottlenecks without costly infrastructure changes. This study advances disaster modeling by bridging disaster development with dynamic evacuation, offering a replicable framework for similar environments. Full article

With the continuous improvement of China’s power grid, safety issues in substation operation and maintenance have become increasingly prominent. However, the existing electrical proximity early-warning devices are inadequate for the complex environments of substations, highlighting the urgent need to develop new electrical proximity early-warning technologies. Based on the safety needs of substation operators, this paper proposes an electrical proximity early-warning method that integrates ‘electric field + distance’. It combines MEMS electric field test technology with ultrasonic ranging technology and designs a double-criterion electrical proximity early-warning device. Based on the COMSOL 6.0 finite-element electric field simulation and the construction safety specification for substation equipment, a multistage electric-field early-warning threshold has been reasonably formulated. A field test conducted at a 220 kV substation demonstrates that this device can issue alerts for various electrical proximity threat levels of the circuit breaker within 0.1 s, which is faster and more accurate than existing commercial electrical proximity early-warning devices. The double-criterion early-warning system minimizes the risk of missed alarms during multi-distance measurements. Additionally, its flexible warning threshold accommodates the increasingly complex operational requirements of substations. Full article

The paper presents a study of an intelligent system for personnel positioning, transport, and equipment monitoring in the mining industry using convolutional neural network (CNN) and OpenPose technology. The proposed framework operates through a three-stage pipeline: OpenPose-based skeleton extraction from surveillance video streams, capturing 18 key body joints at 30fps; multimodal feature fusion, combining skeletal key points and proximity sensor data to achieve environmental context awareness and obtain relevant feature values; and hierarchical pose alert, using attention-enhanced bidirectional LSTM (trained on 5000 annotated fall instances) for fall warning. The experiment conducted demonstrated that the combined use of the aforementioned technologies allows the system to determine the location and behavior of personnel, calculate the distance to hazardous areas in real time, and analyze personnel postures to identify possible risks such as falls or immobility. The system’s capacity to track the location of vehicles and equipment enhances operational efficiency, thereby mitigating the risk of accidents. Additionally, the system provides real-time alerts, identifying abnormal behavior, equipment malfunctions, and safety hazards, thus promoting enhanced mine management efficiency, improved safe working conditions, and a reduction in accidents. Full article

This study introduces ESPCN-YOLO, an innovative deep learning framework designed to enhance the detection accuracy of Personal Protective Equipment (PPE) under challenging conditions, including low-light environments, long-distance scenarios, and small object detection. The proposed system integrates a YOLOv8-based object detection model with an Efficient Sub-Pixel Convolutional Neural Network (ESPCN) to perform real-time super-resolution enhancement on low-resolution footage. The framework was trained on a custom dataset containing 21,750 annotated images categorized into four PPE classes: helmets, shoes, vests, and persons. Extensive experiments were conducted under varying conditions, including distances ranging from 4 to 14 m, resolutions of 640 × 480 and 1920 × 1080, and brightness levels adjusted from −90% to +70%. The results demonstrate that integrating an ESPCN (3×) with YOLOv8 significantly improves detection accuracy, particularly for small objects and poorly illuminated environments. The model achieved a mean average precision (mAP@0.5) of 0.922 and a stringent mAP@0.5:0.95 of 0.741. Additionally, an automated alert system was implemented to enable real-time PPE compliance monitoring. This study highlights the effectiveness of super-resolution enhancement in increasing detection robustness and provides a practical solution for real-time safety monitoring in industrial environments. Full article

The construction industry faces persistent occupational health and safety challenges, with numerous risks arising from construction sites’ complex and dynamic nature. Accidents frequently result from inadequate safety distances and poorly managed work-er–machine interactions, highlighting the need for advanced safety management solutions. This study develops and validates an innovative hazard warning system that leverages deep learning-based image recognition (YOLOv7) and Internet of Things (IoT) modules to enhance construction site safety. The system achieves a mean average precision (mAP) of 0.922 and an F1 score of 0.88 at a 0.595 confidence threshold, detecting hazards in under 1 s. Integrating IoT-enabled smart wearable devices provides real-time monitoring, delivering instant hazard alerts and personalized safety warnings, even in areas with limited network connectivity. The system employs the DIKW knowledge management framework to extract, transform, and load (ETL) high-quality labeled data and optimize worker and machinery recognition. Robust feature extraction is performed using convolutional neural networks (CNNs) and a fully connected approach for neural network training. Key innovations, such as perspective projection coordinate transformation (PPCT) and the security assessment block module (SABM), further enhance hazard detection and warning generation accuracy and reliability. Validated through extensive on-site experiments, the system demonstrates significant advancements in real-time hazard detection, improving site safety, reducing accident rates, and increasing productivity. The integration of IoT enhances scalability and adaptability, laying the groundwork for future advancements in construction automation and safety management. Full article

This paper presents an alarm system and teleoperation control framework, comparing ROS 1 and ROS 2 within a local network to mitigate the risk of robots failing to reach their goals during autonomous navigation. Such failures can occur when the robot moves through irregular terrain, becomes stuck on small steps, or approaches walls and obstacles without maintaining a safe distance. These issues may arise due to a combination of technical, environmental, and operational factors, including inaccurate sensor data, sensor blind spots, localization errors, infeasible path planning, and an inability to adapt to unexpected obstacles. The system integrates a web-based graphical interface developed using frontend frameworks and a joystick for real-time monitoring and control of the robot’s localization, velocity, and proximity to obstacles. The robot is equipped with RGB-D and tracking cameras, a 2D LiDAR, and odometry sensors, providing detailed environmental data. The alarm system provides sensory feedback through visual alerts on the web interface and vibration alerts on the joystick when the robot approaches walls, faces potential collisions with objects, or loses stability. The system is evaluated in both simulation (Gazebo) and real-world experiments, where latency is measured and sensor performance is assessed for both ROS 1 and ROS 2. The results demonstrate that both systems can operate effectively in real time, ensuring the robot’s safety and enabling timely operator intervention. ROS 2 offers lower latency for LiDAR and joystick inputs, making it advantageous over ROS 1. However, camera latency is higher, suggesting the need for potential optimizations in image data processing. Additionally, the platform supports the integration of additional sensors or applications based on user requirements. Full article

We measured the alert distance (AD), flight-initiation distance (FID), buffer distance (BD), and distance fled (DF) of Himalayan marmots (Marmota himalayana) from four populations experiencing human disturbances of the same persistence but different intensities when subjected to varied stimuli (a running or walking man with or without a leashed dog and a dog alone). We analyzed the effects of different stimuli on the AD, FID, BD, and DF of marmots from each population and the relationship among the AD, FID, and DF to illustrate the escape strategy diversification of the studied marmots for different human disturbances when disturbed by varied stimuli. We found that intra-population diversification emerged when the marmots were threatened by different stimuli. The AD and FID were shorter when an individual was walking toward than when he was running toward the focal marmots. A man with a leashed dog as a stimulus produced a similar result to that of a man alone. Nevertheless, no diversification emerged when a single dog was the threat, and all three distances triggered due to the dog were significantly shorter than those triggered due to a man alone (walking or running) or a man with a leashed dog approaching the marmots. Inter-population diversification also emerged when the marmots from the four populations were disturbed by the same stimulus: when threatened by an individual or a man with a leashed dog, their escape behavior was determined by the intensity of the disturbance. The changes in the AD and FID were similar across all four populations, with the two distances increasing with the decrease in disturbance intensity, but the DF showed no significant variation across all the four areas. No significant inter-population diversification emerged when the marmots were threatened by a single dog. These diversifications may result from the different levels of habituation of marmots to human disturbances and the different sizes and, consequently, visibilities of humans and dogs. Full article

Birds and other wildlife are negatively affected by many anthropogenic activities, including human recreational activities, which are often not considered in area planning. Here, I present factors affecting the flight initiation distance (FID)—the distance to an approaching human at which birds flee—for 1075 different flocks of waterbirds. The FID varied greatly between groups of birds and species. For some bird groups and species, the FID was longer in rural areas than in urban areas and increased with flock size and with disturbance from canoeing. In addition to the differences in FID between species and groups of species, there are two important conclusions from this study: (1) a graphical relationship between the proportion of birds that flee at different distances from an approaching person gives more information than mean or median FID values and should be used by nature managers, and (2) the FID should be investigated in each area before mitigating actions or new constructions are decided, considering all the factors affecting it. A global database with a mixture of FID values from a huge number of areas is valuable for some purposes but can be misleading for individuals in a specific area. Full article

As maritime transportation and human activities at sea continue to grow, ensuring the safety of offshore infrastructure has become an increasingly pressing research focus. However, traditional high-precision sensor systems often involve prohibitive costs, and the Automatic Identification System (AIS) faces signal loss or data manipulation problems, highlighting the need for a complementary, affordable, and reliable supplemental solution. This study introduces a monocular vision-based safety monitoring framework for offshore infrastructures. By combining advanced computer vision techniques such as Grounded SAM and horizon-based self-calibration, the proposed framework achieves accurate vessel detection, instance segmentation, and distance estimation. The model integrates open-vocabulary object detection and zero-shot segmentation, achieving high performance without additional training. To demonstrate the feasibility of the framework in practical applications, we conduct several experiments on public datasets and couple the proposed algorithms with the Leaflet.js and WebRTC libraries to develop a web-based prototype for real-time safety monitoring, providing visualized information and alerts for offshore infrastructure operators in our case study. The experimental results and case study suggest that the framework has notable advantages, including low cost, convenient deployment with minimal maintenance, high detection accuracy, and strong adaptability to diverse application conditions, which brings a supplemental solution to research on offshore infrastructure safety. Full article

Background and Objectives: During the COVID-19 pandemic, many countries implemented lockdowns and social distancing measures, which may delay the early diagnosis of colorectal cancer (CRC). This study aims to review the impact of the pandemic on the diagnosis and treatment outcomes of CRC. Materials and Methods: Patients who underwent colonoscopy or surgery for CRC were included. The study was divided into the pre-COVID-19 (January 2019–January 2020), early COVID-19 (February–May 2020), recovery (June–December 2020), and heightened alert (January–December 2021) periods. Cox regression was used to model the waiting time to colonoscopy. Multivariable logistic regression identified associations between time periods and incidence of CRC diagnosed. The characteristics and outcomes of the surgical procedures that were performed were compared across the time periods. Results: A total of 18,662 colonoscopies and 1462 surgical procedures were performed in the study period. Compared to the pre-COVID-19 period, there was a longer time to colonoscopy during the recovery (HR: 0.91; 95% CI: 0.87, 0.94) and heightened alert periods (HR: 0.88; 95% CI 0.85, 0.91). The early COVID-19 (OR: 1.36; 95% CI: 1.04, 1.77) and recovery (OR: 1.20; 95% CI: 1.01, 1.43) periods were associated with higher odds of diagnosing CRC. Compared to the pre-COVID-19 period, there was a higher proportion of ASA 4 patients (4.3% vs. 1.3%; p < 0.001) and stage 4 CRC patients (22.2% vs. 16.9%; p = 0.001) that required surgery during the heightened alert period. Similarly, there was a higher proportion of emergency surgeries (22% vs. 13.3%; p = 0.002); diverting stomas (13.5% vs. 10.5%; p = 0.005), and Hartmann’s procedures (4.4% vs. 0.4%; p = 0.001) performed during the heightened alert period. Conclusions: The pandemic was associated with a higher proportion of metastatic CRC patients requiring surgery. Healthcare policies should facilitate early cancer screening, diagnosis, and treatment to reduce cancer-related morbidity for future pandemics. Full article

Edge processing is a trend in developing new technologies that leverage Artificial Intelligence (AI) without transmitting large volumes of data to centralized processing services. This technique is particularly relevant for security applications where there is a need to reduce the probability of intrusion or data breaches and to decentralize alert systems. Although drone detection has received great research attention, the ability to identify helicopters expands the spectrum of aerial threats that can be detected. In this work, we present the development of a perimeter alert system that integrates AI and multiple sensors processed at the edge. The proposed system can be integrated into a LoRa or LoRaWAN network powered by solar energy. The system incorporates a PDM microphone based on an Arduino Nano 33 BLE with a trained model to identify a drone or a UH-60 from an audio spectrogram to demonstrate its functionality. It is complemented by two PIR motion sensors and a microwave sensor with a range of up to 11 m. Additionally, the DC magnetic field is measured to identify possible sensor movements or changes caused by large bodies, and a configurable RGB light signal visually indicates motion or sound detection. The monitoring system communicates with a second MCU integrated with a LoRa or LoRaWAN communication module, enabling information transmission over distances of up to several kilometers. The system is powered by a LiPo battery, which is recharged using solar energy. The perimeter alert system offers numerous advantages, including edge processing for enhanced data privacy and reduced latency, integrating multiple sensors for increased accuracy, and a decentralized approach to improving security. Its compatibility with LoRa or LoRaWAN networks enables long-range communication, while solar-powered operation reduces environmental impact. These features position the perimeter alert system as a versatile and powerful solution for various applications, including border control, private property protection, and critical infrastructure monitoring. The evaluation results show notable progress in the acoustic detection of helicopters and drones under controlled conditions. Finally, all the original data presented in the study are openly available in an OSF repository. Full article

The construction industry is considered one of the most hazardous industries. The accidents associated with construction equipment are a leading cause of fatalities in the U.S., with one-quarter of all fatalities in the construction industry due to equipment-related incidents, including collisions, struck-by events, and rollovers. While close collaboration among multiple equipment and humans is common, conventional collision alert mechanisms for equipment usually rely on distance sensors with static thresholds, often resulting in too many false alarms, causing drivers’ ignorance. Considering the collaborative operation scenario, this research proposes and develops a dynamic-threshold alert system by recognizing hazardous events based on the types of nearby objects with their orientation or postures and their distances to the system carrier equipment based on image-based recognition and Sim2Real techniques. Two experiments were conducted, and the results show that the system successfully reduced a large number of false near-collision alarms for the collaboration scenarios. Although the accuracy of object recognition and image-based distance estimation is feasible for practical use, it is also easily degraded in the self-obstruction scenario or for equipment with large and movable parts due to incorrect recognition of the bounding boxes of the target objects. Full article

Infrared small target detection technology has been widely applied in the defense sector, including applications such as precision targeting, alert systems, and naval monitoring. However, due to the small size of their targets and the extended imaging distance, accurately detecting drone targets in complex infrared environments remains a considerable challenge. Detecting drone targets accurately in complex infrared environments poses a substantial challenge. This paper introduces a novel model that integrates edge characteristics with multi-scale feature fusion, named Edge-Guided Feature Pyramid Networks (EG-FPNs). This model aims to capture deep image features while simultaneously emphasizing edge characteristics. The goal is to resolve the problem of missing target information that occurs when Feature Pyramid Networks (FPNs) perform continuous down-sampling to obtain deeper semantic features. Firstly, an improved residual block structure is proposed, integrating multi-scale convolutional feature extraction and inter-channel attention mechanisms, with significant features being emphasized through channel recalibration. Then, a layered feature fusion module is introduced to strengthen the shallow details in images while fusing multi-scale image features, thereby strengthening the shallow edge features. Finally, an edge self-fusion module is proposed to enhance the model’s depiction of image features by extracting edge information and integrating it with multi-scale features. We conducted comparative experiments on multiple datasets using the proposed algorithm and existing advanced methods. The results show improvements in the IoU, nIoU, and F1 metrics, while also showcasing the lightweight nature of EG-FPNs, confirming that they are more suitable for drone detection in resource-constrained infrared scenarios. Full article