Search Results (2,045)

This study presents the first systematic evaluation of the capability of the Surface Water and Ocean Topography (SWOT) satellite Level-2 High Rate Pixel Cloud (L2_HR_PIXC) product for retrieving topography in reservoir drawdown zones under varying terrain conditions in arid and semi-arid regions. Three representative reservoirs in southern Xinjiang, China—characterized by plain, canyon, and pocket-shaped canyon morphologies—were selected to establish a terrain-dependent validation framework. A novel multi-feature clustering strategy integrating elevation and radar backscatter coefficients was explored to reduce the misclassification of wet mudflats as water pixels in the PIXC product, aiming to improve DEM accuracy in reservoir drawdown zones. Based on this framework, multi-cycle SWOT-derived digital elevation models (DEMs) were generated and quantitatively evaluated against high-resolution unmanned aerial vehicle (UAV) Light Detection and Ranging (LiDAR) DEMs. Results demonstrate a strong terrain dependency in SWOT-derived elevation accuracy. In low-relief environments, sub-meter accuracy is achieved, with the root mean square error (RMSE) below 0.25 m, confirming the suitability of SWOT for high-precision monitoring. However, errors increase significantly in steep and complex terrains, reaching up to ±6 m, primarily due to interferometric decorrelation, geometric distortion, and slope-induced biases. Despite these limitations, multi-temporal observations exhibit generally similar spatial error patterns across terrains, indicating reasonable repeatability under the tested conditions. This study reveals the performance boundaries of SWOT-derived DEMs in dynamic land–water transition zones and provides a robust methodological framework for improving DEM extraction in similar environments. The findings contribute to advancing the application of SWOT data in hydrological monitoring and geomorphological analysis at regional scales. Full article

Medical thermography is a noninvasive, contactless imaging technique that captures spatial temperature distributions across the human body, providing insights into vascular function, inflammation, metabolism, physiological regulation, and aging. Recently, machine learning has been increasingly utilized to analyze thermographic data for disease screening, functional assessment, and biomarker identification. However, the existing literature is fragmented, with varied clinical applications, feature-engineering strategies, and predictive modeling frameworks, often lacking a focus on interpretability and the reliable identification of clinically relevant thermal markers. This review offers a structured overview of explainable machine learning in medical thermography, emphasizing thermal feature representation, model interpretability, and biomarker discovery. It categorizes thermographic features into pixel-based representations, region-wise statistical descriptors, texture measures, and deep latent features. Additionally, it evaluates conventional machine learning and deep learning methods for classification, regression, and risk assessment tasks. The review pays special attention to interpretable learning strategies, such as feature importance analysis, surrogate explanation models, saliency-based visualization, and Shapley-value-based methods, which can enhance transparency and confidence in model outputs. Key challenges are critically discussed, including imaging variability, limited dataset sizes, weak protocol standardization, class imbalance, generalizability, and the gap between predictive performance and clinical trust. Overall, this review synthesizes current advancements, identifies major research gaps, and outlines future directions for developing trustworthy machine learning frameworks in medical thermography and enhancing interpretable thermal biomarker discovery. Full article

To address the difficulty of finely classifying complex optical remote sensing images, this paper innovatively proposes a new image classification method based on quantum statistics (QS) inspired by quantum physics. Each pixel in the image is regarded as a fermion, which is one of the fundamental particles in quantum systems. The energy of the energy level where fermions are located is described using the negative logarithm of the distribution that the spectrum of the pixel follows. The Fermi-Dirac distribution, a quantum statistics model used to describe the complex occupation pattern of energy levels by fermions, is employed to characterize the membership relationship between pixels and classes, instead of traditional distance measures and probability measures. Then, the cost function guiding the convergence of classification is defined based on free energy, which is used to describe whether a system is in a state of thermal equilibrium according to energy, temperature, and entropy. To minimize the free energy, the derivative method and the simulated annealing algorithm are adopted to estimate the optimal solution for model parameters. The proposed method can describe complex features more effectively, obtain fine classification results, and overcome the curse of dimensionality in high-dimensional image classification. Finally, the feasibility and effectiveness are verified through qualitative and quantitative analysis of multispectral and hyperspectral image classification experiments. Full article

Thin-film interference in photoresist stacks can become a significant source of uncertainty in lithographic focus metrology, particularly when high measurement stability is required. To evaluate this effect, a Fresnel-based multilayer reflection model is used to analyze the optical response of the resist stack and to guide the selection of dual-wavelength illumination. On this basis, a dual-wavelength optical triangulation system is developed for focus metrology in 350 nm lithography, with signal acquisition performed by a linear charge-coupled device (LCCD). Rather than improving precision by reducing detector pitch, the system employs a two-stage sub-pixel localization strategy in which template matching provides coarse spot localization and weighted centroid interpolation refines the final position within localized calculation windows, keeping the computational cost manageable. A covariance-based uncertainty analysis predicts a total root-mean-square uncertainty of 27.23 nm. Prototype experiments were performed on a bare silicon wafer to establish the intrinsic performance of the instrument before introducing process-dependent optical effects. Under these conditions, the system achieved a vertical resolution of 10 nm, a repeatability of 35 nm, and a stability of 13.16 nm. The additional uncertainty expected under resist-coated-wafer conditions was assessed separately through the thin-film model. These results verify the baseline capability of the proposed system and support the feasibility of the dual-wavelength strategy for focus metrology in 350 nm lithography. Full article

To address the strong reliance on manual operations and the low efficiency of egg shape index (ESI) phenotyping in layer breeding, this study proposed an ESI measurement method based on an improved YOLOv12n-seg model. Ghost Bottleneck modules were introduced into the backbone to reduce model complexity. In addition, a boundary-aware loss combining Binary cross entropy (BCE), Dice, and Boundary Loss was designed to improve mask quality. Based on the segmentation results generated by YOLO-Ghost, principal component analysis was employed to extract the orientation and scale of the principal axes of the segmented regions. The major and minor axes of the pixel-level masks were then obtained, and their ratio was used as the measured ESI value. Compared with YOLOv12n-seg, YOLO-Ghost reduced the number of model parameters and computational cost by 39.86% and 17.58%, respectively, while increasing the frame rate by 40.91%. The model achieved an mAP@0.50–0.95 of 92.10%, BF1 of 86.28%, and BIoU of 74.99%. Compared with other instance segmentation models, YOLO-Ghost achieved a precision of 99.96%, a recall of 99.69%, and a detection speed of 454.55 f/s. For ESI estimation, the predicted values showed good agreement with manual measurements, with an R² of 0.8184, MAE of 0.03219, and RMSE of 0.03681. The results indicate that the proposed method can achieve non-contact, automated, and accurate measurement of ESI, and provides technical support for high-throughput automated phenotypic data collection in layer breeding. Full article

This article presents RAW-FABRID (RAW FABric Image Dataset), a publicly available annotated dataset for raw fabric defect detection using computer vision techniques. It addresses a major limitation in textile inspection, where reliance on private datasets hinders objective methodological comparisons. RAW-FABRID was acquired using a custom-built inspection machine equipped with controlled LED illumination and a line-scan camera. The dataset includes grayscale fabric images collected from several manufacturers to ensure variability in textures and patterns. It comprises 709 high-resolution images (1792 × 1024 pixels), including both defect-free and defective samples. To maximize reusability, data are provided in two complementary formats: high-resolution images (cropped to remove peripheral acquisition artifacts) for global analysis, and a patch-based organization following the widely adopted MVTec Anomaly Detection benchmark structure. The latter divides images into 256 × 256 pixel patches for direct machine learning integration. Crucially, the dataset is accompanied by comprehensive metadata (CSV) and precise COCO-formatted annotations (JSON) for both subsets, ensuring full traceability and supporting object detection and semantic segmentation. The dataset is publicly available through Mendeley Data, enabling reproducible research and objective benchmarking of defect detection algorithms. Full article

The interaction between crossflows from sprinkler nozzles and airflow is crucial for engineering applications, particularly affecting the efficiency of sprayed areas. This study investigates the deformation of a continuously injected conical water spray subjected to horizontal airflow, using a planar laser imaging method as a visualisation technique. Experiments were conducted in a wind tunnel at a constant water pressure of 0.2 MPa and four airflow rates (0.1, 0.2, 0.4, and 0.6 m³·s⁻¹) to systematically vary the air-to-water momentum ratio. A grayscale-based analysis method was developed using a per-pixel Look-Up Table (LUT), enabling indirect assessment of droplet concentrations and spray structure. This approach allowed for a detailed examination of changes in the spray cone shape under flowing air. By assessing the water spray across three vertical planes intersecting the spray cone, it became possible to calculate lateral area and cone volume at different air-to-water mass flow ratios. The spray formation region exposed to airflow exhibited larger cone volumes than those with minimal airflow. The changes in apparent spray angles for the tested nozzles were determined to characterize the cone shape. The apparent spray angle varies systematically with the air-to-water mass flow ratio, confirming the dominant role of aerodynamic forces. These findings improve the understanding of spray behavior under crossflow and provide a basis for validating numerical models of air–water interactions. Full article

Most existing low-dimensional chaotic maps suffer from a limited dynamical complexity and dynamic degradation, which restrict their effectiveness in image encryption. To address this issue, a novel three-dimensional chaotic map (3D-TCM) was constructed to improve dynamical complexity and stability, and its superiority was verified through a dynamical analysis. Based on these advantages, a plaintext-related image encryption scheme was designed by combining bit-level permutation and S-box-based diffusion. The experimental results show that the proposed scheme achieved high information entropy, a low pixel correlation, and desirable NPCR and UACI values, and successfully passed NIST SP800-22 statistical tests, demonstrating a strong resistance to differential attacks and overall robustness. Full article

Inspection inside small-diameter pipelines is difficult because the narrow interior space limits the field of view of onboard cameras. Even when a crack is successfully detected, it may still appear near the image boundary rather than in a suitable position for observation. To address this issue, this study proposes a vision-guided active crack alignment framework for small-diameter pipe inspection robots. The proposed framework uses a YOLOv5s detector to identify the crack region and extract the center of the detected bounding box. The positional difference between the crack center and the image center is defined as the image-plane alignment error. After low-pass filtering, this error is converted into actuator-side reference input through a pixel-to-motor mapping, and a PID-based closed-loop controller is used to regulate a local camera adjustment mechanism so that the detected crack region moves toward the image center. The framework is evaluated mainly through simulation, including controller comparison, different initial offset conditions, parameter sensitivity analysis, robustness tests under visual fluctuation and mapping uncertainty, and an ablation study. The controller comparison shows that all tested PID-based controllers achieve stable convergence, while the fuzzy PID controller provides the best overall performance among the tested cases in terms of settling time, steady-state error, and RMS error. The framework also remains stable under different crack positions and moderate uncertainty conditions. In addition, a preliminary laboratory-scale physical consistency test is conducted to examine whether the convergence tendency observed in simulation can also be reproduced under real visual feedback and actuator response. The preliminary physical results show a convergence tendency consistent with the simulation trend, thereby providing initial support for the practical implementability of the proposed detection-driven alignment concept. Complete integration with an in-pipe robot platform and validation under realistic pipe environments remain future work. Full article

This study investigates how land use changes upstream and the Kamal Khan Dam have reshaped patterns of water allocation and intensified hydropolitical tensions in the Helmand/Hirmand Transboundary River Basin (HTRB). Remote sensing and deep learning techniques were employed to analyze land use changes from 2012 to 2024. After radiometric and atmospheric corrections were applied to Landsat imagery, pixel-based classification was performed using a Deep Belief Network model. Additionally, hydrological changes were analyzed using Sentinel-2 data, with attention paid to the diversion of water flows toward the Godzareh Depression. The classification results revealed considerable expansion of agricultural land downstream of the Kajaki Dam, primarily at the expense of rangeland, forest, and barren land. Moreover, sentinel imagery confirmed that, following the commissioning of the Kamal Khan Dam in 2021, systematically diverted Helmand/Hirmand hydrosystem flows toward the Godzareh Depression. Also, this study applied the painted water framework to analyze the evolution of hydropolitical dynamics in the HTRB across two critical periods, 1954–1980 and 2010–2025. The analysis revealed a fundamental transformation from green water-dominated natural flow regimes to infrastructure-controlled systems characterized by concurrent increases in both yellow water (controlled but not immediately consumed) and red water (controlled and consumed). The Kamal Khan Dam’s operationalization represents a pivotal inflection point, dramatically expanding Afghanistan’s yellow water reserves. This dual expansion of controlled water categories, empirically documented through satellite imagery, confirms the emergence of negative hydrohegemony in the basin. Consequently, Iran’s position has transitioned from a historically stable painted water class to one characterized by critical dependency. Full article

Small unmanned aerial vehicle (UAV)-borne distributed tomographic synthetic aperture radar (TomoSAR) systems offer flexible baseline configurations and low deployment cost, making them attractive for rapid and high-resolution three-dimensional (3D) reconstruction. However, the distance between adjacent channels placed on different UAVs is relatively large due to the flight safety spacing considerations. This leads to high sidelobes in the elevation point spread function (PSF) within the reconstruction range. Meanwhile, atmospheric turbulence may cause UAVs to deviate from their predefined trajectories, making it difficult to suppress sidelobes through baseline optimization. Large baselines may also introduce spatial decorrelation between channels, which gives rise to random phase noise in the interferometric phase and further aggravates elevation ambiguity by increasing the sidelobe level of the PSF. To address this problem, this paper proposes an elevation ambiguity resolution method based on neighborhood-adaptive elevation priors. In the proposed method, a window function is constructed from reconstruction results of neighboring pixels and incorporated into the reconstruction process to suppress the interference caused by high sidelobes. In this way, the probability of correct target reconstruction is improved. The effectiveness and robustness of the proposed method are validated using both simulations and real measured data. Experimental results obtained with a C-band small UAV-borne distributed TomoSAR system show that the proposed method effectively suppresses ambiguity and enables ambiguity-free reconstruction of target buildings. Statistical analysis further demonstrates that the number of ambiguous points produced by the proposed algorithm is only one-fifth of that produced by the conventional OMP method. Full article

Early detection of malignancy is imperative, yet existing diagnostic approaches struggle to identify small peripheral lesions. This study evaluated a novel imaging modality, heat diffusion analysis, to assess its ability to differentiate between malignant and normal lung tissue. Considering that lung cancer is the leading cause of cancer-related mortality worldwide, lung tumors were induced in mice in a preclinical ex vivo model to evaluate the proposed technology. The HTOScan System was used to analyze the thermal characteristics of 60 sites from excised lungs, including normal and abnormal regions. The algorithm classified pixels as high- or low-risk for malignancy. The HTOScan System demonstrated a high accuracy of 97%, with 94% sensitivity and 98% specificity compared to the gold standard of histopathology. The technology successfully differentiated abnormal from normal tissue ex vivo based on differences in thermal diffusivity. This proof-of-concept study suggests that combining heat diffusion imaging techniques with machine learning algorithms could enable the HTOScan System to identify malignant lesions accurately with high confidence. The technique shows promise as a real-time decision support tool for cancer detection, pending further in vivo validation. This novel functional-imaging approach could improve the identification of peripheral lesions and the guidance of biopsies during bronchoscopy. Full article

Power-system reliability is crucial for sustainable development, but large-scale, long-term monitoring remains challenging. Existing nighttime light (NTL)-based outage detection methods often rely on fixed thresholds or prior information, limiting cross-regional application. To address this, we develop an adaptive thresholding framework using daily NASA Black Marble data. Observations are grouped by view angle to mitigate radiometric instability, and a per-pixel dynamic baseline is constructed from high-radiance statistics, enabling robust anomaly detection without prior outage timing. From the detected anomalies, we formulate a population-weighted NTL power reliability index (NTPRI) to quantify regional electricity service reliability. Validation across six diverse outage events yields an F1 score of 0.807. National-scale analysis shows NTPRI correlates significantly with the World Bank’s System Average Interruption Duration Index (SAIDI). The derived Light Anomaly Rate (LAR) further supports pixel-level frequency analysis. Together, this framework provides a transferable remote-sensing tool for large-scale power-reliability assessment in data-scarce regions, supporting disaster impact evaluation and energy vulnerability analysis. Full article

This study addresses the problem of robust video-based tracking of laboratory rats in open-field and Y-maze experiments under challenging acquisition conditions, including non-uniform illumination, low contrast, and heterogeneous recording setups. Existing approaches based on classical image processing or deep learning often fail to maintain stable localization under such conditions or require large, annotated datasets. We propose a hybrid tracking framework that combines an improved motion–appearance voting mechanism with consistency-constrained optimization for open-field experiments, together with a comparative deep learning-based detection strategy for Y-maze analysis. The proposed method introduces (i) adaptive dual-threshold motion extraction, (ii) directionally constrained temporal validation, and (iii) a robustness-driven fusion of motion and appearance cues. Experimental results demonstrate that the proposed approach achieves reliable tracking with a maximum localization error below 10 pixels under severe illumination variations. In the Y-maze scenario, a comparative evaluation of multiple detectors (YOLOv5, YOLOv9, YOLO12, Faster R-CNN) highlights the trade-off between accuracy and inference time, with YOLOv9 providing the best balance. The main contribution consists of enabling robust behavioral quantification in low-quality experimental conditions using limited training data, bridging the gap between classical tracking robustness and deep learning flexibility. Full article

This paper proposes a finite-time target tracking control for multi-rotor unmanned aerial vehicles (UAVs) based on a vision-inertial-stabilized platform. To address the challenge of stable and accurate moving target tracking, the sliding mode control (SMC) technique is used to overcome limitations of conventional control algorithms, such as poor robustness and slow convergence speed. First, by computing the pixel deviation between the target and the image center, a kinematic model of the tracking target is established. Then, by introducing homogeneous system theory into the sliding mode surface design, a non-singular fast integral terminal sliding mode control (NFITSMC) is designed for target tracking via regulating the rotational angular acceleration of dual actuators in the vision inertial stabilized platform, thereby driving the pixel deviation to converge to zero in a finite time. Strict theoretical analysis is given to prove the finite-time stability and robustness of the closed-loop control system. Furthermore, simulation results demonstrate that the proposed method maintains higher tracking accuracy than SMC, ISMC, and TSMC. Full article