Search Results (181)

Distributed acoustic sensing (DAS), which can provide dense spatial and temporal measurements using optical fibers, is quickly becoming critical for large-scale infrastructure monitoring. However, anomaly detection in DAS data is still challenging owing to the spatial correlations between sensing channels and nonlinear temporal dynamics. Recent approaches often disregard the explicit sensor layout and instead handle DAS data as two-dimensional images or flattened sequences, eliminating the spatial topology. This work proposes GraphDiffusion, a novel generative anomaly-detection model that combines a conditional denoising diffusion probabilistic model (DDPM) and a graph neural network (GNN) to overcome these limitations. By treating each channel as a graph node and building edges based on Euclidean proximity, the GNN explicitly models the spatial arrangement of DAS sensors, allowing the network to capture local interchannel dependencies. The conditional DDPM uses iterative denoising to model the temporal dynamics of standard signals, enabling the system to detect deviations without the need for anomalies. The performance evaluations based on real-world DAS datasets reveal that GraphDiffusion achieves 98.2% and 98.0% based on the area under the curve (AUC) of the F1-score at K different levels (F1_K-AUC), an AUC of receiver operating characteristic (ROC) at K different levels (ROC_K-AUC), outperforming other comparative models. Full article

Research on image sensor (IS)-based visible light positioning systems has attracted widespread attention. However, when the receiver is tilted or under a single LED, the positioning system can only achieve two-dimensional (2D) positioning and requires the assistance of inertial measurement units (IMU). When the LED is not captured or decoding fails, the system’s positioning error increases further. Thus, we propose a novel three-dimensional (3D) visible light positioning system based on image sensors for various environments. Specifically, (1) we use IMU to obtain the receiver’s state and calculate the receiver’s 2D position. Then, we fit the height–size curve to calculate the receiver’s height, avoiding the coordinate iteration error in traditional 3D positioning methods. (2) When no LED or decoding fails, we propose a firefly-assisted unscented particle filter (FA-UPF) algorithm to predict the receiver’s position, achieving high-precision dynamic positioning. The experimental results show that the system positioning error under a single LED is within 10 cm, and the average positioning error through FA-UPF under no light source is 6.45 cm. Full article

Background: Artificial intelligence (AI) and machine learning (ML) have been reshaping maternal, fetal, neonatal, and reproductive healthcare by enhancing risk prediction, diagnostic accuracy, and operational efficiency across the perinatal continuum. However, no comprehensive synthesis has yet been published. Objective: To conduct a scoping review of reviews of AI/ML applications spanning reproductive, prenatal, postpartum, neonatal, and early child-development care. Methods: We searched PubMed, Embase, the Cochrane Library, Web of Science, and Scopus through April 2025. Two reviewers independently screened records, extracted data, and assessed methodological quality using AMSTAR 2 for systematic reviews, ROBIS for bias assessment, SANRA for narrative reviews, and JBI guidance for scoping reviews. Results: Thirty-nine reviews met our inclusion criteria. In preconception and fertility treatment, convolutional neural network-based platforms can identify viable embryos and key sperm parameters with over 90 percent accuracy, and machine-learning models can personalize follicle-stimulating hormone regimens to boost mature oocyte yield while reducing overall medication use. Digital sexual-health chatbots have enhanced patient education, pre-exposure prophylaxis adherence, and safer sexual behaviors, although data-privacy safeguards and bias mitigation remain priorities. During pregnancy, advanced deep-learning models can segment fetal anatomy on ultrasound images with more than 90 percent overlap compared to expert annotations and can detect anomalies with sensitivity exceeding 93 percent. Predictive biometric tools can estimate gestational age within one week with accuracy and fetal weight within approximately 190 g. In the postpartum period, AI-driven decision-support systems and conversational agents can facilitate early screening for depression and can guide follow-up care. Wearable sensors enable remote monitoring of maternal blood pressure and heart rate to support timely clinical intervention. Within neonatal care, the Heart Rate Observation (HeRO) system has reduced mortality among very low-birth-weight infants by roughly 20 percent, and additional AI models can predict neonatal sepsis, retinopathy of prematurity, and necrotizing enterocolitis with area-under-the-curve values above 0.80. From an operational standpoint, automated ultrasound workflows deliver biometric measurements at about 14 milliseconds per frame, and dynamic scheduling in IVF laboratories lowers staff workload and per-cycle costs. Home-monitoring platforms for pregnant women are associated with 7–11 percent reductions in maternal mortality and preeclampsia incidence. Despite these advances, most evidence derives from retrospective, single-center studies with limited external validation. Low-resource settings, especially in Sub-Saharan Africa, remain under-represented, and few AI solutions are fully embedded in electronic health records. Conclusions: AI holds transformative promise for perinatal care but will require prospective multicenter validation, equity-centered design, robust governance, transparent fairness audits, and seamless electronic health record integration to translate these innovations into routine practice and improve maternal and neonatal outcomes. Full article

The automated detection and classification of immature macauba (Acrocomia aculeata) fruits is critical for improving post-harvest processing and quality control. In this study, we present a comparative evaluation of two state-of-the-art YOLO architectures, YOLOv11x and YOLOv12x, trained on the newly constructed VIC01 dataset comprising 1600 annotated images captured under both background-free and natural background conditions. Both models were implemented in PyTorch and trained until the convergence of box regression, classification, and distribution-focal losses. Under an IoU (intersection over union) threshold of 0.50, YOLOv11x and YOLOv12x achieved an identical mean average precision (mAP₅₀) of 0.995 with perfect precision and recall or TPR (true positive rate). Averaged over IoU thresholds from 0.50 to 0.95, YOLOv11x demonstrated superior spatial localization performance (mAP_50–95 = 0.973), while YOLOv12x exhibited robust performance in complex background scenarios, achieving a competitive mAP_50–95. Inference throughput averaged 3.9 ms per image for YOLOv11x and 6.7 ms for YOLOv12x, highlighting a trade-off between speed and architectural complexity. Fused model representations revealed optimized layer fusion and reduced computational overhead (GFLOPs), facilitating efficient deployment. Confusion-matrix analyses confirmed YOLOv11x’s ability to reject background clutter more effectively than YOLOv12x, whereas precision–recall and F1-score curves indicated both models maintain near-perfect detection balance across thresholds. The public release of the VIC01 dataset and trained weights ensures reproducibility and supports future research. Our results underscore the importance of selecting architectures based on application-specific requirements, balancing detection accuracy, background discrimination, and computational constraints. Future work will extend this framework to additional maturation stages, sensor fusion modalities, and lightweight edge-deployment variants. By facilitating precise immature fruit identification, this work contributes to sustainable production and value addition in macauba processing. Full article

With the increase in the penetration rate of distributed sources and loads, the sensor monitoring data is increasing dramatically. Power grid maintenance services require a rapid response in power quality data analysis. To achieve a rapid response and highly accurate classification of power quality disturbances (PQDs), this paper proposes an efficient classification algorithm for PQDs based on Lissajous trajectory (LT) and a lightweight DenseNet, which utilizes the concept of Lissajous curves to construct an ideal reference signal and combines it with the original PQD signal to synthesize a feature trajectory with a distinctive shape. Meanwhile, to enhance the ability and efficiency of capturing trajectory features, a lightweight L-DenseNet skeleton model is designed, and its feature extraction capability is further improved by integrating an attention mechanism with L-DenseNet. Finally, the LT image is input into the fusion model for training, and PQD classification is achieved using the optimally trained model. The experimental results demonstrate that, compared with current mainstream PQD classification methods, the proposed algorithm not only achieves superior disturbance classification accuracy and noise robustness but also significantly improves response speed in PQD classification tasks through its concise visualization conversion process and lightweight model design. Full article

The downwelling light sensor (DLS) is the industry-standard solution for generating UAV-based digital orthophoto maps (DOMs). Current mainstream DLS correction methods primarily rely on angle compensation. However, due to the temporal mismatch between the DLS sampling intervals and the exposure times of multispectral cameras, as well as external disturbances such as strong wind gusts and abrupt changes in flight attitude, DLS data often become unreliable, particularly at UAV turning points. Building upon traditional angle compensation methods, this study proposes an improved correction approach—FIM-DC (Fitting and Interpolation Model-based Data Correction)—specifically designed for data collection under clear-sky conditions and stable atmospheric illumination, with the goal of significantly enhancing the accuracy of reflectance retrieval. The method addresses three key issues: (1) field tests conducted in the Qingpu region show that FIM-DC markedly reduces the standard deviation of reflectance at tie points across multiple spectral bands and flight sessions, with the most substantial reduction from 15.07% to 0.58%; (2) it effectively mitigates inconsistencies in reflectance within image mosaics caused by anomalous DLS readings, thereby improving the uniformity of DOMs; and (3) FIM-DC accurately corrects the spectral curves of six land cover types in anomalous images, making them consistent with those from non-anomalous images. In summary, this study demonstrates that integrating FIM-DC into DLS data correction workflows for UAV-based multispectral imagery significantly enhances reflectance calculation accuracy and provides a robust solution for improving image quality under stable illumination conditions. Full article

Perovskite bionic eyes have emerged as highly promising candidates for photodetection applications to their wide-angle imaging capabilities, high external quantum efficiency(EQE), and low-cost fabrication and integration. Since their initial exploration in 2015, significant advancements have been achieved in this field, with their EQE reaching 27%. Nevertheless, intrinsic challenges such as the oxidation susceptibility of perovskites and difficulties in curved surface growth hinder their further development. Addressing these issues necessitates a comprehensive and systematic understanding of the preparation mechanisms for hemispherical perovskite, as well as the development of effective mitigation strategies. In this review, a review published provides a detailed overview of the research progress in hemispherical perovskite photodetectors, with a particular focus on the fundamental properties and fabrication pathways of hemispherical perovskites. Furthermore, various strategies to enhance the performance of hemispherical perovskite and overcome preparation challenges are thoroughly discussed. Finally, existing challenges and perspectives are presented to further advance the development of eco-friendly hemispherical perovskite. Full article

Agriculture in India supports millions of livelihoods and is a major force behind economic expansion. Challenges in modern agriculture depend on environmental factors (such as soil quality and climate variability) and biotic factors (such as pests and diseases). These challenges can be addressed by advancements in technology (such as sensors, internet of things, communication, etc.) and data-driven approaches (such as machine learning (ML) and deep learning (DL)), which can help with crop yield and sustainability in agriculture. This study introduces an innovative deep neural network (DNN) approach for identifying leaf diseases in paddy crops at an early stage. The proposed neural network is a hybrid DL model comprising feature extraction, channel attention, inception with residual, and classification blocks. Channel attention and inception with residual help extract comprehensive information about the crops and potential diseases. The classification module uses softmax to obtain the score for different classes. The importance of each block is analyzed via an ablation study. To understand the feature extraction ability of the modules, extracted features at different stages are fed to the SVM classifier to obtain the classification accuracy. This technique was experimented on eight classes with 7857 paddy crop images, which were obtained from local paddy fields and freely available open sources. The classification performance of the proposed technique is evaluated according to accuracy, sensitivity, specificity, F1 score, MCC, area under curve (AUC), and receiver operating characteristic (ROC). The model was fine-tuned by setting the hyperparameters (such as batch size, learning rate, optimizer, epoch, and train and test ratio). Training, validation, and testing accuracies of 99.91%, 99.87%, and 99.49%, respectively, were obtained for 20 epochs with a learning rate of 0.001 and sgdm optimizer. The proposed network robustness was studied via an ablation study and with noisy data. The model’s classification performance was evaluated for other agricultural data (such as mango, maize, and wheat diseases). These research outcomes can empower farmers with smarter agricultural practices and contribute to economic growth. Full article

Arc scanning synthetic aperture radar (ArcSAR) can achieve high-resolution panoramic imaging and retrieve submillimeter-level deformation information. To monitor buildings in a city scenario, ArcSAR must be lightweight; have a high resolution, a mid-range (around a hundred meters), and low power consumption; and be cost-effective. In this study, a novel high-resolution wide-beam single-chip millimeter-wave (mmwave) ArcSAR system, together with an imaging algorithm, is presented. First, to handle the non-uniform azimuth sampling caused by motor motion, a high-accuracy angular coder is used in the system design. The coder can send the radar a hardware trigger signal when rotated to a specific angle so that uniform angular sampling can be achieved under the unstable rotation of the motor. Second, the ArcSAR’s maximum azimuth sampling angle that can avoid aliasing is deducted based on the Nyquist theorem. The mathematical relation supports the proposed ArcSAR system in acquiring data by setting the sampling angle interval. Third, the range cell migration (RCM) phenomenon is severe because mmwave radar has a wide azimuth beamwidth and a high frequency, and ArcSAR has a curved synthetic aperture. Therefore, the fourth-order RCM model based on the range-Doppler (RD) algorithm is interpreted with a uniform azimuth angle to suit the system and implemented. The proposed system uses the TI 6843 module as the radar sensor, and its azimuth beamwidth is 64°. The performance of the system and the corresponding imaging algorithm are thoroughly analyzed and validated via simulations and real data experiments. The output image covers a 360° and 180 m area at an azimuth resolution of 0.2°. The results show that the proposed system has good application prospects, and the design principles can support the improvement of current ArcSARs. Full article

This manuscript proposes a non-contact approach to characterize geometrical configuration and strain rate effects using digital image correlation (DIC). The non-contact DIC technique allows more robust and accurate material property assessment than conventional in-contact gauges, especially under dynamic loading conditions. This study first demonstrates DIC-based strain measuring accuracy in quasi-static coupon tests with two geometrical configurations. In comparison to the conventional method, DIC measures a wider range of strain up to the final fracture while eliminating geometric constraints typically imposed on test specimens. This study further extends the DIC measurement in dynamic material property tests, i.e., the split-Hopkinson bar test. The direct strain measurement by DIC presents enhanced accuracy compared to the conventional method, as the latter overestimates the strain results from remotely installed strain gauges. The deformation analysis explains the discrepancy in strain measurement at different sensor locations. The strain rate effects on the stress–strain curve and fracture strain are evaluated on two materials, i.e., aluminum alloy 5083 and S550 steel. The proposed DIC approach enables more convenient and robust measurement of strains, which facilitates the material property evaluation under various geometrical configurations and strain rates. Full article

A sufficient number of sensors installed in all structural components is a prerequisite for obtaining the full-field temporal–spatial displacement and is essential for large-scale structure health monitoring. In this paper, a novel lightweight vision-based temporal–spatial deflection measurement method is proposed to measure the full-field temporal–spatial displacement of slender structures. First, the geometric and mechanical properties of slender members are introduced as the priori information to vision-based measurement. Then, a slice template matching model is proposed by deploying a one-dimensional template matching model in every pixel column of each image frame, based on traditional digital image correlation (DIC) method. An indoor experiment was carried out to verify the proposed method, and experiment results show that measurement precision of STMM agrees well with the theory and the laser ranger, with a maximum measurement error of 0.03 pixels and the root-mean-square error (RMSE) of 0.052 mm, for transient beam deflection curve; with the correlation coefficient and coefficient of determination of 0.9994 and 0.9986, for dynamic deflection–time history curves at the middle-span point. Finally, further investigation reveals that brightness inconstancy is the source of STMM measurement error. Full article

Unmanned aerial vehicles (UAVs) equipped with hyperspectral hardware systems are widely used in urban planning and land classification. However, hyperspectral sensors generate large volumes of data that are rich in both spatial and spectral information, making its efficient processing in resource-constrained devices challenging. While transformers have been widely adopted for hyperspectral image classification due to their global feature extraction capabilities, their quadratic computational complexity limits their applicability for resource-constrained devices. To address this limitation and enable the real-time processing of hyperspectral data on UAVs, we propose a lightweight multi-head MambaOut with a CosTaylorFormer (LMHMambaOut-CosTaylorFormer). First, 3D-2D CNN is used to extract both spatial and spectral shallow features from hyperspectral images. Following this, one branch employs a linear transformer, CosTaylorFormer, to extract global spectral information. More specifically, we propose CosTaylorFormer with a cosine function, adjusting the weights based on the spectral curve distribution, which is more conducive to establishing long-distance spectral dependencies. Meanwhile, compared with other linearized transformers, the CosTaylorFormer we propose better improves model performance. For the other branch, we propose multi-head MambaOut to extract global spatial features and enhance the network classification effect. Moreover, a dynamic information fusion strategy is proposed to adaptively fuse spatial and spectral information. The proposed network is validated on four datasets (IP, WHU-Longkou, SA, and PU) and compared with several models, demonstrating its superior classification accuracy; however, the number of model parameters is only 0.22 M, thus achieving better balance between model complexity and accuracy. Full article

The water status of plants is affected by abiotic and biotic environmental factors and influences the growth and yield formation of crops. Assessment of the leaf water content (LWC) of grapevine using hyperspectral imaging (1000–2500 nm) was investigated under controlled conditions for its potential to study the effects of the downy mildew pathogen Plasmopara viticola on LWC of host tissue in compatible and incompatible interactions. A calibration curve was established for the relationship between LWC and the Normalized Difference Leaf Water Index (NDLWI₁₉₃₇) that uses spectral information from the water absorption band and NIR for normalization. LWC was significantly lower for abaxial than for adaxial leaf sides, irrespective of grapevine genotype and health status. Reflecting details of leaf anatomy, vascular tissue exhibited effects reverse to intercostal areas. Effects of P. viticola on LWC coincided with the appearance of first sporangia on the abaxial side and increased during further pathogenesis. Continuous water loss ultimately resulted in tissue death, which progressed from the margins into central leaf areas. Tiny spots of brown leaf tissue related to the reaction of partial resistant cultivars could be monitored only at the sensor’s highest spatial resolution. Proximal sensing enabled an unprecedented spatial resolution of leaf water content in host–pathogen interactions and confirmed that resistance reactions may produce a combination of dead and still-living cells that enable the development of biotrophic P. viticola. Full article

This study presents a novel method for detecting sleep posture changes—specifically tossing and turning—by monitoring variations in humidity using an array of humidity sensors embedded at regular intervals within a memory-foam pillow. Unlike previous approaches that rely primarily on temperature or pressure sensors, our method leverages the observation that humidity fluctuations are more pronounced during movement, enabling the more sensitive detection of posture changes. We demonstrate that dynamic patterns in humidity data correlate strongly with physical motion during sleep. To identify these transitions, we applied the Pruned Exact Linear Time (PELT) algorithm, which effectively segmented the time series based on abrupt changes in humidity. Furthermore, we converted humidity fluctuation curves into image representations and employed a transfer-learning-based model to classify sleep postures, achieving accurate recognition performance. Our findings highlight the potential of humidity sensing as a reliable modality for non-invasive sleep monitoring. In this study, we propose a novel method for detecting tossing and turning during sleep by analyzing changes in humidity captured by a linear array of sensors embedded in a memory foam pillow. Compared to temperature data, humidity data exhibited more significant fluctuations, which were leveraged to track head movement and infer sleep posture. We applied a rolling smoothing technique and quantified the cumulative deviation across sensors to identify posture transitions. Furthermore, the PELT algorithm was utilized for precise change-point detection. To classify sleep posture, we converted the humidity time series into images and implemented a transfer learning model using a Vision Transformer, achieving a classification accuracy of approximately 96%. Our results demonstrate the feasibility of a sleep posture analysis using only humidity data, offering a non-intrusive and effective approach for sleep monitoring. Full article

Stretch blow molding (SBM) is widely utilized in industrial applications, yet the deformation characteristics of materials during this process are intricate and challenging to precisely articulate. To accurately forecast the stress–strain response of polyethylene terephthalate (PET) in SBM, a hybrid Artificial Neural Network (ANN)-based constitutive model has been developed. The model has been created by combining a physical-based function for capturing the small-strain behavior in parallel with an ANN-based model for capturing the temperature-dependent large-strain nonlinear viscoelastic behavior. The architecture of the ANN has been designed to ensure stability in a load-controlled scenario, thus making it suitable for use in FEA simulations of stretch blow molding. Data for training the model have been generated by a new semi-automatic experimental rig which is able to produce 850 stress–strain curves over a wide range of process conditions (temperature range 95–115 °C and strain rates ranging from 1/s to 100/s) directly from blowing preforms using a combination of high-speed video, digital image correlation and sensors for pressure and force. The model has already been implemented in the commercial FEA package Abaqus via a VUMAT subroutine, with its performance validated by comparing the prediction of the evolution of preform shape during blowing vs. high-speed images. In conclusion, the developed hybrid ANN model, when integrated into Abaqus, offers a more accurate simulation of SBM processes, contributing to improved design efficiency and product quality. Full article