Search Results (82)

Reliable detection of personnel is critical for situational awareness and life-saving interventions during indoor fire rescue operations, where dense smoke rapidly obscures visibility and compromises conventional vision systems. Visible-light cameras fail under such conditions due to severe Mie scattering, while thermal infrared (TIR) imaging—though capable of penetrating smoke—often lacks the fine-grained texture needed to distinguish human forms from background clutter. Furthermore, practical deployment of multimodal sensors is hindered by spatial misalignment between modalities, which degrades fusion efficacy and detection accuracy. To address these challenges, this paper proposes GLAFC-YOLO (Global-Local Alignment and Frequency-aware Cross-attention Fusion), a dual-stream multimodal detection framework specifically designed for personnel localization in smoke-obscured indoor fires. GLAFC-YOLO fuses near-infrared (NIR) and TIR imagery through three novel components: (1) a Global-Local Feature Alignment Subnet (GL-FAS) that rectifies geometric misalignment across modalities; (2) a Modality-Adaptive Frequency Channel Attention (MA-FCA) module that enhances complementary smoke-penetrating thermal signatures and NIR texture cues in the frequency domain; and (3) a Confidence-Aware Transposed Cross-Attention (CA-TCA) mechanism that suppresses smoke-induced artifacts and restores weakened human-centric features. Evaluated on a newly collected multimodal dataset of indoor fire scenarios with annotated personnel, GLAFC-YOLO achieves substantial improvements over the baseline YOLOv11 architecture. Specifically, it achieves Recall improvements of 43.2% and 0.5% compared to unimodal NIR and TIR baselines, respectively. In addition, it achieves improvements of 37.4% and 3.9% in $mA{P}_{50}$ and 17.3% and 17.0% in $mA{P}_{50–95}$. Experimental results indicate that GLAFC-YOLO outperforms competitive models and reduces personnel miss rates, demonstrating its robustness and readiness for real-world fire-rescue assistance. Full article

Estimating the position and orientation of a camera with respect to an observed scene remains a fundamental problem in computer vision, particularly in calibration procedures and multi-sensor vision systems. This paper revisits the planar Perspective–n–Point (PnP) problem with emphasis on rotation representation, initialization strategy, and optimization behavior. We propose the PnP-ProCay78 algorithm, which combines analytical elimination of translation via quadratic reconstruction error with nonlinear least-squares minimization of projection residuals in Cayley parameter space. A deterministic initialization scheme based on canonical directions of the reconstruction matrix eliminates the need for spectral search over the full solution space. Experimental evaluation on heterogeneous datasets acquired from high-resolution RGB cameras and low-resolution thermal cameras demonstrates that the proposed method achieves reprojection accuracy comparable to state-of-the-art OpenCV implementations such as SQPnP and IPPE. Convergence analysis in Cayley space reveals stable and rapidly contracting optimization trajectories, with consistent behavior across sensors of significantly different resolution and noise characteristics. The results indicate that a carefully chosen rotation parameterization combined with a transparent optimization framework can yield competitive numerical performance while maintaining geometric interpretability and structural simplicity. Full article

The feather coverage of a laying hen is an important indicator of both its productivity and welfare. Conventional manual feather scoring procedures are laborious, subjective, and stressful for the hens. Thermography offers a modern alternative to addressing these problems. Thermal cameras capture radiative heat loss, which is comparatively greater Classification from featherless areas. Studies have been conducted to establish a standard temperature range that correlates to specific featherless areas. However, such temperature-based approaches have been inconsistent with each other. In contrast, this study used deep learning techniques to automatically assess dorsal feather scores using thermal images. Thermal images (n = 1575) of the dorsal body of cage-free laying hens with varying degrees of feather damage were captured. Manual feather scoring was performed, classifying the image into a feather score (0–2) according to the increasing severity of feather loss. A total of 1222 images were selected, filtering out images of lower quality. Two types of computer vision models, a classification model and an object detection model, were trained and evaluated. A custom convolutional neural network (CNN) was trained to classify thermal images into feather score categories. Additionally, we trained and optimized You Only Look Once (YOLO) object detection models to detect areas of feather damage and predict the feather score. The CNN model achieved an overall accuracy of 0.81, with high precision for severe feather loss. The YOLO-based object detection model was optimum using YOLO11n, which achieved a precision of 0.81, a recall of 0.73 and a mean average precision (mAP) at 0.5 intersection over union (IoU) of 0.84. Results show the potential of combining thermal imaging with deep learning techniques to perform objective, automatic, and scalable feather scoring procedures. Future studies should focus on data diversity, multiple part scoring, and semantic segmentation for robust performance. Full article

The proliferation of air travel demand necessitates innovative solutions to enhance passenger experience while optimizing airport operational efficiency. This paper presents the pilot-scale implementation and evaluation of an AI-powered service robot ecosystem integrated with thermal cameras and 5G wireless connectivity at Athens International Airport. The system addresses critical challenges in passenger flow management through real-time crowd analytics, congestion detection, and personalized robotic assistance. Eight strategically deployed thermal cameras monitor passenger movements across check-in areas, security zones, and departure entrances while employing privacy-by-design principles through thermal imaging technology that reduces personally identifiable information capture. A humanoid service robot, equipped with Robot Operating System navigation capabilities and natural language processing interfaces, provides real-time passenger assistance including flight information, wayfinding guidance, and congestion avoidance recommendations. The wi.move platform serves as the central intelligence hub, processing video streams through advanced computer vision algorithms to generate actionable insights including passenger count statistics, flow rate analysis, queue length monitoring, and anomaly detection. Formal trial evaluation conducted on 10 April 2025, with extended operational monitoring from April to June 2025, demonstrated strong technical performance with application round-trip latency achieving 42.9 milliseconds, perfect service reliability and availability ratings of one hundred percent, and comprehensive passenger satisfaction scores exceeding 4.3/5 across all evaluated dimensions. Results indicate promising potential for scalable deployment across major international airports, with identified requirements for sixth-generation network capabilities to support enhanced multi-robot coordination and advanced predictive analytics functionalities in future implementations. Full article

Stereo vision is critical for environmental perception in autonomous driving, but faces challenges in accuracy and stability under extreme automotive temperature cycles. This study addresses environment-induced deformation in tri-camera imaging systems through material and structural optimization to enhance ranging stability. Using finite element analysis (Abaqus), we evaluated three aluminum alloys (AL6063-T6, AL6061, AL7075-T6), a heterogeneous structure (AL6063-T6/AL1060), and a honeycomb design under operational temperatures (−40 °C, 25 °C, 95 °C). Results show AL6063-T6 exhibits superior thermal stability, minimizing optical axis offset (δ ≈ 0.134° vs. 0.143° for AL7075-T6). The AL6063-T6/AL1060 heterogeneous structure further reduced deformation (δ ≈ 0.133°), while the honeycomb design increased offset (δ ≈ 0.145°). The experimental results also show that AL6063-T6 exhibits better deformation resistance than AL6061 and AL7075-T6, which helps reduce camera ranging errors and improve the stability of stereo vision imaging. The experimental results are consistent with the finite element analysis, validating the effectiveness of the finite element analysis for the camera material optimization design. These findings demonstrate that material selection and heterogeneous structural design significantly mitigate environment-induced deformation, improving tri-camera ranging accuracy and imaging stability for automotive applications. Full article

The traditional inference of thermal comfort relies mainly on either questionnaire surveys or invasive physiological signal monitoring. However, the use of these methods in real time is limited and they have a low accuracy; furthermore, they can cause an inconvenience to the daily work and life of indoor personnel. With the development of intelligent building technology, non-intrusive technology based on video analyses has gradually become a research hotspot. Not only does this type of technology avoid the limitations of traditional methods, but it can also be used to dynamically monitor thermal comfort. At present, the established and relatively complete non-intrusive recognition methods usually rely on additional equipment or cameras with specific angles, which limits their deployment and application in a wider range of scenarios. Therefore, in order to improve the non-intrusive prediction accuracy of the thermal comfort level of indoor personnel, it is necessary to establish a non-intrusive indoor personnel thermal comfort inference model. This study designed a cross-modal knowledge-distillation-based thermal adaptive behavior-recognition model. In order to avoid the difficulties of terminal deployment caused by the large model and the time-consuming nature of optical flow estimation, a multi-teacher network model was used to transfer the knowledge of different modes to a single student model. This reduced the number of model parameters and the computational complexity while improving the recognition accuracy. The experimental results show that the proposed vision-based thermal adaptation behavior-recognition model can non-invasively and accurately identify the thermal adaptation behavior of indoor personnel, which can not only improve the comfort of indoor environments, but also enable the intelligent adjustment of HVAC systems. Full article

The parameterization of vegetation indices (VIs) is crucial for sustainable irrigation and horticulture management, specifically for urban green infrastructure (GI) management. However, the constraints of roadside traffic, motor and industrially related pollution, and potential public vandalism compromise the efficacy of conventional in situ monitoring systems. The shortcomings of prevalent satellites, UAVs, and manual/automated sensor measurements and monitoring systems have already been reviewed. This research proposes a novel urban GI monitoring system based on an integration of gas exchange and various VIs obtained from computer vision algorithms applied to data acquired from three novel sources: (1) Integrated gas sensor data using nine different volatile organic compounds using an electronic nose (E-nose), designed on a PCB for stable performance under variable environmental conditions; (2) Plant growth parameters including effective leaf area index (LAIe), infrared index (Ig), canopy temperature depression (CTD) and tree water stress index (TWSI); (3) Meteorological data for all measurement campaigns based on wind velocity, air temperature, rainfall, air pressure, and air humidity conditions. To account for spatial and temporal data acquisition variability, the integrated cameras and the E-nose were mounted on a vehicle roof to acquire information from 172 Elm trees planted across the Royal Parade, Melbourne. Results showed strong correlations among air contaminants, ambient conditions, and plant growth status, which can be modelled and optimized for better smart irrigation and environmental monitoring based on real-time data. Full article

Heat stress is one of the main welfare and productivity problems faced by dairy cattle in Mediterranean climates. The main objective of this work is to predict heat stress in livestock from shade-seeking behavior captured by computer vision, combined with some climatic features, in a completely non-invasive way. To this end, we evaluate two soft computing algorithms—Random Forests and Neural Networks—clarifying the trade-off between accuracy and interpretability for real-world farm deployment. Data were gathered at a commercial dairy farm in Titaguas (Valencia, Spain) using overhead cameras that counted cows in the shade every 5–10 min during summer 2023. Each record contains the shaded-cow count, ambient temperature, relative humidity, and an exact timestamp. From here, three thermal indices were derived: the current THI, the previous-night mean THI, and the day-time accumulated THI. The resulting dataset covers 75 days and 6907 day-time observations. To evaluate the models’ performance a 5-fold cross-validation is also used. The results show that both soft computing models outperform a single Decision Tree baseline. The best Neural Network (3 hidden layers, 16 neurons each, learning rate $={10}^{-3}$) reaches an average RMSE of $14.78$, while a Random Forest (10 trees, depth $=5$) achieves $14.97$ and offers the best interpretability. Daily error distributions reveal a median RMSE of $13.84$ and confirm that predictions deviate less than one hour from observed shade-seeking peaks. Although the dataset came from a single farm, the results generalized well within the observed range. However, the models could not accurately predict the exact number of cows in the shade. This suggests the influence of other variables not included in the analysis (such as solar radiation or wind data), which opens the door for future research. Full article

Thermography was first applied to breast cancer detection in the 1950s but fell out of favor among radiologists due to inconsistent and inconclusive findings in the following decades. Studies conducted in the 21st century using new-generation thermal cameras and computer vision techniques, particularly artificial intelligence, have reported sensitivity and specificity values comparable to those of mammography. However, most radiologists, being unaware of these results, still believe this technique is ineffective, and medical societies advise against using it, even as an adjunct to mammography. In this paper we review recent studies and discuss whether the recommendations of scientific societies are still valid in the light of new evidence. We also propose some ideas for standardizing breast thermography studies that could help make this technique acceptable to the radiology community. Full article

Urban thermal environment is directly linked to the health and comfort of local residents, as well as energy consumption. Drone-based thermal infrared image acquirement provides an efficient and flexible way of assessing urban heat distribution, thereby supporting climate-resilient and sustainable urban development. Here, we present an advanced approach that utilizes the thermal infrared camera mounted on the drone for high-resolution building wall temperature measurement and achieves centimeter accuracy. According to the binocular vision theory, the three-dimensional (3D) reconstruction of thermal infrared images is first conducted, and then the two-dimensional building wall temperature is extracted. Real-world validation shows that our approach can measure the wall temperature within a 5 °C error, which confirms the reliability of this approach. The field measurement of Yuquanting in Xiong’an New Area China during three time periods, i.e., morning (7:00–8:00), noon (13:00–14:00) and evening (18:00–19:00), was used as a case study to demonstrate our approach. The results show that during the heating season, the building wall temperature was the highest at noon time and the lowest in evening time, which were mostly caused by solar radiation. The highest wall temperature at noon time was 55 °C, which was under direct sun radiation. The maximum wall temperature differences were 39 °C, 55 °C, and 20 °C during morning, noon and evening time, respectively. The lighter wall coating color tended to have a lower temperature than the darker wall coating color. Beyond this application, this approach has potential in future autonomous thermal environment measuring systems as a foundational element. Full article

Many researchers focus on improving reproductive health in sows and ensuring successful breeding by accurately identifying the optimal time of ovulation through estrus detection. One promising non-contact technique involves using computer vision to analyze temperature variations in thermal images of the sow’s vulva. However, variations in camera distance during dataset collection can significantly affect the accuracy of this method, as different distances alter the resolution of the region of interest, causing pixel intensity values to represent varying areas and temperatures. This inconsistency hinders the detection of the subtle temperature differences required to distinguish between estrus and non-estrus states. Moreover, failure to maintain a consistent camera distance, along with external factors such as atmospheric conditions and improper calibration, can distort temperature readings, further compromising data accuracy and reliability. Furthermore, without addressing distance variations, the model’s generalizability diminishes, increasing the likelihood of false positives and negatives and ultimately reducing the effectiveness of estrus detection. In our previously proposed methodology for estrus detection in sows, we utilized YOLOv8 for segmentation and keypoint detection, while monocular depth estimation was used for camera calibration. This calibration helps establish a functional relationship between the measurements in the image (such as distances between labia, the clitoris-to-perineum distance, and vulva perimeter) and the depth distance to the camera, enabling accurate adjustments and calibration for our analysis. Estrus classification is performed by comparing new data points with reference datasets using a three-nearest-neighbor voting system. In this paper, we aim to enhance our previous method by incorporating the mean pixel intensity of the region of interest as an additional factor. We propose a detailed four-step methodology coupled with two stages of evaluation. First, we carefully annotate masks around the vulva to calculate its perimeter precisely. Leveraging the advantages of deep learning, we train a model on these annotated images, enabling segmentation using the cutting-edge YOLOv9 algorithm. This segmentation enables the detection of the sow’s vulva, allowing for analysis of its shape and facilitating the calculation of the mean pixel intensity in the region. Crucially, we use monocular depth estimation from the previous method, establishing a functional link between pixel intensity and the distance to the camera, ensuring accuracy in our analysis. We then introduce a classification approach that differentiates between estrus and non-estrus regions based on the mean pixel intensity of the vulva. This classification method involves calculating Euclidean distances between new data points and reference points from two datasets: one for “estrus” and the other for “non-estrus”. The classification process identifies the five closest neighbors from the datasets and applies a majority voting system to determine the label. A new point is classified as “estrus” if the majority of its nearest neighbors are labeled as estrus; otherwise, it is classified as “non-estrus”. This automated approach offers a robust solution for accurate estrus detection. To validate our method, we propose two evaluation stages: first, a quantitative analysis comparing the performance of our new YOLOv9 segmentation model with the older U-Net and YOLOv8 models. Secondly, we assess the classification process by defining a confusion matrix and comparing the results of our previous method, which used the three nearest points, with those of our new model that utilizes five nearest points. This comparison allows us to evaluate the improvements in accuracy and performance achieved with the updated model. The automation of this vital process holds the potential to revolutionize reproductive health management in agriculture, boosting breeding success rates. Through thorough evaluation and experimentation, our research highlights the transformative power of computer vision, pushing forward more advanced practices in the field. Full article

Conveyor belts are critical in various industries, particularly in the barrier eddy current separator systems used in recycling processes. However, hidden issues, such as belt misalignment, excessive heat that can lead to fire hazards, and the presence of sharp or irregularly shaped materials, reduce operational efficiency and pose serious threats to the health and safety of personnel on the production floor. This study presents an intelligent monitoring and protection system for barrier eddy current separator conveyor belts designed to safeguard machinery and human workers simultaneously. In this system, a thermal camera continuously monitors the surface temperature of the conveyor belt, especially in the area above the magnetic drum—where unwanted ferromagnetic materials can lead to abnormal heating and potential fire risks. The system detects temperature anomalies in this critical zone. The early detection of these risks triggers audio–visual alerts and IoT-based warning messages that are sent to technicians, which is vital in preventing fire-related injuries and minimizing emergency response time. Simultaneously, a machine vision module autonomously detects and corrects belt misalignment, eliminating the need for manual intervention and reducing the risk of worker exposure to moving mechanical parts. Additionally, a line-scan camera integrated with the YOLOv11 AI model analyses the shape of materials on the conveyor belt, distinguishing between rounded and sharp-edged objects. This system enhances the accuracy of material separation and reduces the likelihood of injuries caused by the impact or ejection of sharp fragments during maintenance or handling. The YOLOv11n-seg model implemented in this system achieved a segmentation mask precision of 84.8 percent and a recall of 84.5 percent in industry evaluations. Based on this high segmentation accuracy and consistent detection of sharp particles, the system is expected to substantially reduce the frequency of sharp object collisions with the BECS conveyor belt, thereby minimizing mechanical wear and potential safety hazards. By integrating these intelligent capabilities into a compact, cost-effective solution suitable for real-world recycling environments, the proposed system contributes significantly to improving workplace safety and equipment longevity. This project demonstrates how digital transformation and artificial intelligence can play a pivotal role in advancing occupational health and safety in modern industrial production. Full article

This paper proposes an automated system integrating infrared thermography (IRT) and computer vision for air leak detection and localization in end-of-line (EOL) testing stations. This system consists of (1) a leak tester for detection and quantification of leaks, (2) an infrared camera for real-time thermal image acquisition; and (3) an algorithm for automatic leak localization. The python-based algorithm acquires thermal frames from the camera’s streaming video, identifies potential leak regions by selecting a region of interest, mitigates environmental interferences via image processing, and pinpoints leaks by employing pixel intensity thresholding. A closed circuit with an embedded leak system simulated relevant leakage scenarios, varying leak apertures (ranging from 0.25 to 3 mm), and camera–leak system distances (0.2 and 1 m). Results confirmed that (1) the leak tester effectively detected and quantified leaks, with larger apertures generating higher leak rates; (2) the IRT performance was highly dependent on leak aperture and camera–leak system distance, confirming that shorter distances improve localization accuracy; and (3) the algorithm localized all leaks in both lab and industrial environments, regardless of the camera–leak system distance, mostly achieving accuracies higher than 0.7. Overall, the combined system demonstrated great potential for long-term implementation in EOL leakage stations in the manufacturing sector, offering an effective and cost-effective alternative for manual inspections. Full article

Remote sensing visual-light spectral (VIS) maps provide stable and rich features for geo-localization. However, it still remains a challenge to make use of VIS map features as localization references at night. To construct a cross-day-and-night localization system for long-duration UAVs, this study proposes a visual–inertial integrated localization system, where the visual component can register both RGB and infrared camera images in one unified VIS map. To deal with the large differences between visible and thermal images, we inspected various visual features and utilized a pre-trained network for cross-domain feature extraction and matching. To obtain an accurate position from visual geo-localization, we demonstrate a localization error compensation algorithm with considerations about the camera attitude, flight height, and terrain height. Finally, the inertial and dual-vision information is fused with a State Transformation Extended Kalman Filter (ST-EKF) to generate long-term, drift-free localization performance. Finally, we conducted actual long-duration flight experiments with altitudes ranging from 700 to 2400 m and flight distances longer than 344.6 km. The experimental results demonstrate that the proposed method’s localization error is less than 50 m in its RMSE. Full article

Novel view synthesis and 3D reconstruction have been extensively studied. Three-dimensional Gaussian Splatting (3DGS) has gained popularity due to its rapid training and real-time rendering capabilities. However, RGB imaging is highly dependent on ideal illumination conditions. In low-light situations such as at night or in the presence of occlusions, RGB images often suffer from blurred contours or even complete failure in imaging, which severely restricts the application of 3DGS in such scenarios. Thermal imaging technology, on the other hand, serves as an effective complement. Thermal images are solely influenced by heat sources and are immune to illumination conditions. This unique property enables them to clearly identify the contour information of objects in low-light environments. Nevertheless, thermal images exhibit significant limitations in presenting texture details due to their sensitivity to temperature variations rather than surface texture features. To capitalize on the strengths of both, we propose thermal geometrically accurate Gaussian Splatting (TGA-GS), a novel Gaussian Splatting model. TGA-GS is designed to leverage RGB and thermal information to generate high-quality meshes in low-light conditions. Meanwhile, given low-resolution thermal images and low-light RGB images as inputs, our method can generate high-resolution thermal and RGB images from novel viewpoints. Moreover, we also provide a real thermal imaging dataset captured with a handheld thermal infrared camera. This not only enriches the information content of the images but also provides a more reliable data basis for subsequent computer vision tasks in low-light scenarios. Full article