Search Results (2,318)

Wildfires represent a major environmental hazard with significant impacts on ecosystems, climate, biodiversity, and human activities. The increasing frequency and intensity of wildfire events have highlighted the need for reliable and timely detection techniques based on satellite remote sensing. This study investigates the application of Proper Orthogonal Decomposition (POD) to thermal observations acquired from the Spinning Enhanced Visible and Infrared Imager (SEVIRI) onboard the Meteosat Second Generation (MSG) satellite for wildfire anomaly detection. A wildfire event that occurred on 8 August 2021 in Calabria, Southern Italy, was selected as the primary case study. To assess the consistency of the POD response beyond the primary case, the analysis was further extended to two additional wildfire events, Viggianello–Abate and Pazzano–Montestella, using the 15 × 15 pixel extraction window. Middle Infrared (MIR, 3.9 $\mathsf{\mu }$m) observations collected at 15 min intervals over a complete day were analyzed using four different spatial extraction windows (3 × 3, 15 × 15, 30 × 30, and 45 × 45 pixels). POD was employed to separate dominant background thermal variability from localized fire-induced anomalies. The analysis focused on higher-order POD modes, particularly the 6th, 7th, and 8th modes, which exhibited enhanced sensitivity to wildfire activity. Results showed that POD successfully identified thermal anomalies corresponding to wildfire occurrence times independently detected by the RST-FIRES methodology. The comparison of extraction window sizes revealed that the 15 × 15 pixel window provided the best balance between anomaly enhancement, spatial localization, and noise reduction. Larger windows introduced excessive spatial smoothing and reduced localization capability, whereas the smallest window was more affected by noise. The findings demonstrate the potential of POD as an effective complementary approach for wildfire detection and monitoring using geostationary satellite observations. Full article

Reliable pedestrian perception under low illumination is important for nighttime monitoring in smart built environments. However, visible-light detectors often lose texture and edge information, whereas conventional RGB–infrared fusion may introduce cross-modal noise and discard discriminative cues during scale conversion. This study proposes Multimodal Wavelet–Spectral DETR (MWSD), a frequency-aware RGB–infrared detection framework. MWSD employs a dual-branch Multimodal Fusion Feature Sampling backbone for cross-modal interaction. The Multimodal Frequency-Domain Feature Enhancement (MFFE) module produces input-dependent Fourier modulation within shared detection features, rather than reconstructing a fused image or independently fusing modality-specific spectra. Haar wavelet upsampling and downsampling (HWU and HWD) construct a bidirectional feature pyramid by using frequency components to guide adjacent-level scale conversion, rather than performing image-level wavelet reconstruction. This coordinated design combines residual spectral enhancement with wavelet-guided multi-scale fusion in an end-to-end detector. On LLVIP, MWSD achieves 96.7% $mA{P}_{50}$ and 63.0% $mA{P}_{50:95}$. On M${}^3$FD, it achieves 87.1% and 59.0%, respectively. The model requires 45 ms per 640 × 640 image on an NVIDIA RTX 4090 GPU. These results support frequency-aware multimodal detection as a visual perception approach for nighttime safety monitoring. Full article

Visible–near-infrared (VNIR) hyperspectral imaging provides a non-contact approach for cultural heritage examination. This study presents the design and calibration of a compact finite-conjugate VNIR pushbroom hyperspectral camera for close-range mural imaging. Operating over 400–1000 nm at a nominal working distance of 404 mm, the system provides a mean spectral sampling interval of 4.85 nm and an object-space sampling interval of approximately 82.4 μm/pixel. An integrated calibration workflow was established for wavelength assignment, spectral response characterization, geometric correction, radiometric calibration, and scan synchronization. The experimental results yielded a modulation transfer function (MTF) of 0.34 at the effective detector Nyquist frequency, a mean spectral response function full width at half maximum (FWHM) of 6.3 nm, a maximum absolute wavelength residual below 0.90 nm, residual smile and keystone errors below 0.3 pixels, a residual radiometric nonuniformity of 0.71%, and a mean signal-to-noise ratio (SNR) of 339. Measurements of Potala Palace mural samples demonstrate the acquisition of spatially detailed, radiometrically corrected hyperspectral data under close-range conditions. Full article

A near-infrared (NIR) light transmission imaging system was constructed using a stereomicroscope equipped with an 860 nm NIR LED light source to evaluate its applicability for observing wood anatomical structures. Species identification of wooden Buddhist statues is important for clarifying their provenance and production period, but such objects require non-destructive examination, and surface darkening from aging and soot deposits often limits observation under visible light. Fifteen wood species used for Buddhist statues and other cultural and architectural properties—eight hardwoods and seven softwoods—were examined, and NIR transmission images of the transverse section were obtained under air-dried and water-saturated conditions. Under air-dried conditions, NIR transmittance differed among species and between heartwood and sapwood, revealing anatomical structures such as vessels, growth-ring boundaries, and resin canals. Under water-saturated conditions, transmittance increased in all species and resin-canal structures became more distinct in softwoods, although some image blurring occurred. These results indicate that NIR transmission observation is effective for non-destructive species identification of wooden cultural properties, and that wood moisture strongly affects the resulting transmitted image. Full article

Accurate shank-length phenotyping of live chickens is hindered by feather occlusion and uncertain endpoint localization in visible images. We developed a two-stage method that fuses registered visible and infrared images before predicting the two measurement endpoints. The fusion network used a U-Net encoder–decoder with residual blocks, Coordinate Attention, and Strip Pooling, trained with a YUV-guided loss. A YOLOv8s-Pose model with Coordinate Attention and a length-related loss then localized the endpoints. The dataset comprised 100 chickens and 1000 paired visible–infrared acquisitions, separated at the individual level into training, validation, and test sets. On the test set, at the chicken level, the reported mean signed difference, mean absolute error, and root mean square error were 0.129 mm, 0.790 mm, and 0.984 mm, respectively. The Pearson correlation coefficient between model-derived and manual reference measurements was 0.992. Compared with either single-modality input, the fused images reduced the mean absolute error and root mean square error. These findings show that complementary texture and thermal-boundary information can support accurate vision-based shank-length measurement under controlled acquisition conditions. Full article

Infrared and visible image fusion (IVIF) aims to preserve infrared thermal targets and visible structural textures in one informative image. Although recent attention-based methods improve cross-modal interaction, their post-fusion refinement remains limited in two aspects: modality-specific channel statistics are no longer explicitly exposed after feature mixing, and repeated attention-based aggregation can smooth spatial responses and weaken high-frequency visible details. To address these issues, this work proposes a lightweight end-to-end IVIF network with two complementary refinement modules. MSG carries out cross-modal style-based recalibration by making use of the joint mean and standard deviation of the two pre-fusion encoder features, so that first- and second-order pre-fusion modality statistics can guide post-fusion channel selection. DGM carries out edge enhancement by constructing a parameter-free Sobel detail prior from source images and learning only a lightweight residual modulation to perform restoration of high-frequency evidence. With only 80,160 trainable parameters, the proposed method achieves the best or tied-best value on three of seven standard fusion-quality metrics on FMB and four of seven on LLVIP, and ablation results further confirm the complementary effects of MSG and DGM. Full article

Large yellow croaker (Larimichthys crocea) is an economically important marine fish, yet rapid and non-destructive discrimination between sea-caught and farmed collection-source groups remains challenging. This study developed a fish-level analytical workflow integrating YOLO11n-seg instance segmentation with visible–near-infrared hyperspectral imaging to extract relative reflectance ratios from six anatomical regions. The analytical cohort comprised 258 unique fish, including 171 sea-caught and 87 farmed individuals. Source categories were assigned according to the original capture or cage-culture collection channels. Fish identity was used as the grouping unit in five-fold internal cross-validation, and classification performance was evaluated based on out-of-fold fish-level predictions. Logistic-regression models achieved AUC values ranging from 0.985 to 1.000 across the six individual anatomical regions. The combined six-region model achieved an accuracy of 0.996 (95% CI, 0.988–1.000) and an AUC of 1.000 (95% CI, 1.000–1.000) within the present cohort. For anatomical-region segmentation, the 20-image validation set contained 140 annotated instances. Bounding-box precision, recall, mAP@0.5, and mAP@0.5–0.95 were 0.976, 0.983, 0.981, and 0.700, respectively, while the corresponding mask metrics were 0.969, 0.976, 0.972, and 0.669. These results indicate that region-specific hyperspectral information can support highly accurate internal discrimination between the two collection-source groups and provide an interpretable basis for characterizing source-associated spectral differences. However, the source labels were not independently verified, potentially influential covariates were incompletely recorded, and no independent external cohort was available. Therefore, the present findings should be interpreted as internally validated collection-source discrimination rather than verified provenance authentication or evidence of external generalizability. Full article

This study investigated how detection distance affects spectral models for soluble solids content (SSC) and firmness evaluation in Korla fragrant pears and provides a reference for calibrating standardized indoor non-destructive detection equipment. Two hundred visually intact fruit samples at the early-ripening stage were collected from the Korla production area in Xinjiang. An FS-640 multispectral camera system equipped with a VS-SWR fixed-focus industrial lens (16 mm focal length, F1.8 maximum aperture, 1/2-inch sensor format) was used to acquire fruit reflectance spectra at seven vertical lens-to-fruit-surface distances of 90, 100, 110, 120, 130, 140, and 150 cm. A 625-pixel region of interest (ROI) was selected using ENVI at an undamaged equatorial or near-equatorial position of each fruit, and the regional mean spectrum was used as the spectral feature of one fruit sample. The sample-set partitioning based on joint X–Y distances (SPXY) algorithm was used to divide the calibration and prediction sets at a 3:1 ratio after outlier removal via a residual-threshold method. Four preprocessing methods, namely LOESS smoothing, standardization, vector normalization, and Savitzky–Golay (SG) smoothing, were compared. Competitive adaptive reweighted sampling (CARS) was performed with 50 Monte-Carlo sampling runs, a maximum of 30 principal components, and 10-fold cross-validation, yielding 99 characteristic wavelengths. Partial least squares regression (PLSR), support vector regression (SVR), random forest (RF), and artificial neural network (ANN) models were then established using identical input variables and sample partitions. Model performance was evaluated using the coefficient of determination for calibration ($R_c^2$), coefficient of determination for prediction ($R_P^2$), root-mean-square error of calibration (RMSEC), root-mean-square error of prediction (RMSEP), relative prediction deviation (RPD), and ratio of performance to interquartile distance (RPIQ). Under the static laboratory acquisition conditions in this work, the SSC model achieved the best prediction performance at 110 cm with SG smoothing ($R_P^2$) = 0.8949, RPD = 3.0633, RPIQ = 5.8661), whereas the firmness model obtained optimal prediction performance at 140 cm with standardization ($R_P^2$) = 0.7460, RPD = 1.9425, RPIQ = 3.2867). Changes in detection distance altered illumination uniformity, effective reflected signal, photon-scattering paths, and background-noise proportion. These effects may partially explain why the chemical-absorption-dominated SSC index and the tissue-scattering-dominated firmness index responded differently to detection distance. The results provide a reference for setting spectral detection parameters for Korla fragrant pears; however, samples were obtained from only a single producing region, harvest season, and maturity stage, and no independent external validation dataset was used. Therefore, the generalization ability of the developed models needs to be further verified using cross-season and cross-orchard sample sets. Full article

Background: The early diagnosis of eye fundus pathologies is crucial, as they may go unnoticed until reaching advanced stages. To offer an improved screening methodology for this purpose, the effectiveness of a conditional variational autoencoder (CVAE) based on multispectral (MS) imaging and operating from the visible to the near-infrared range (416–1213 nm) has been assessed. Methods: A total of 2040 images from 102 patients (66 females, 36 males; aged 19–91 years) were acquired with an MS fundus camera to feed a fine-tuned CVAE for classifying eye fundus as healthy or diseased. The performance of the neural network was assessed for different image resolutions and spectral ranges. Results: The proposed deep generative model showed excellent results, reaching 100% of accuracy, sensitivity and specificity for the set of MS images from 416 nm to 955 nm at maximum resolution (1757 × 1757 pixels). Other instances with different image resolutions and spectral ranges led to good classifications (accuracy between 96% and 98%, sensitivity between 92% and 98%, and specificity between 97% and 100%). The CVAE exhibited robust performance with convergence of the accuracy and loss through the different epochs for training and validation in all instances. Conclusions: This study proves that a CVAE approach based on MS imaging is a highly effective tool for diagnosing eye fundus conditions and could potentially serve as a valuable clinical support tool for screening. The approach performs remarkably well when high spatial resolution MS images ranging from 416 nm to 955 nm are used. This underscores the importance of combining spatial and spectral information, particularly of wavelengths beyond the visible range. Full article

Facial surface temperature provides useful non-contact information for monitoring dairy-cow health, welfare, heat stress, and reproductive status. Infrared thermography can capture thermal information from facial regions such as the eyes, muzzle, nostrils, and ears; however, infrared images often contain weak texture and indistinct anatomical boundaries, which can hinder reliable cow-head localization during mobile robotic inspection. Visible-light images provide richer structural information but do not contain temperature data. This study developed YOLO11-AFE, a lightweight visible–thermal cow-head detection and heterogeneous-camera association method for quadruped inspection robots. The detector incorporates ADown for lightweight downsampling, C3k2_FE for adaptive feature enhancement, and SPPF_ECA for channel-aware multi-scale representation. An improved Hungarian matching algorithm was subsequently used to establish one-to-one correspondences between cow-head detections in synchronized visible-light and infrared pseudo-colour images. Across three independent runs, YOLO11-AFE achieved precision, mAP@0.5, and mAP@0.5:0.95 values of 97.59 ± 0.20%, 96.37 ± 0.24%, and 70.76 ± 0.51%, respectively, on the visible-light test subset, and 96.11 ± 0.24%, 99.07 ± 0.10%, and 91.50 ± 0.40%, respectively, on the infrared subset. The model required 2.14 million parameters and 5.27 GFLOPs, representing reductions of 17.37% and 18.17%, respectively, relative to YOLO11n. The association method achieved an overall accuracy of 98.11% across 371 ground-truth cow-head pairs in the combined validation and test evaluation. TensorRT FP16 deployment on the Jetson Orin NX achieved 36.71 FPS for the complete core processing pipeline. These results demonstrate that YOLO11-AFE provides an accurate and computationally efficient perception front end for future non-contact facial-temperature monitoring using mobile inspection robots. Full article

Deep networks dominate photovoltaic (PV) thermographic fault detection but need large annotated datasets and resist interpretation. We present a lightweight, interpretable infrared workflow combining support-vector-machine (SVM) module/background segmentation from four handcrafted features with an adaptive grid analysis labelling regions as nominal-intensity, high-intensity anomaly or low-intensity anomaly relative to a module-internal reference; the anomaly classes are inspection candidates, not confirmed faults. Evaluation used 21 close-range images of one 20 W module—recorded with the camera’s visible-light edge fusion active, so they are fused infrared/visible frames—and all 596 of a public five-sector UAV dataset. Segmentation against manual masks reached a mean intersection-over-union of 0.64; a feature ablation shows intensity statistics dominate, and an end-to-end Otsu pipeline gives almost the same high-intensity share (4.54% versus 4.50%): the SVM contributes reproducibility—removing the manual segmentation threshold, though not the empirical +48/−60 offsets—not accuracy. High-intensity regions concentrated in the module’s lower half, co-locating with a bus-bar defect known from hardware inspection—suggestive, not validated. The single-module, image-level close-range evaluation is optimistic, and the UAV shares, from a separately trained SVM, illustrate cross-domain application only. Segmentation runs at about 15 images per second on CPU. The method is a relative-intensity thermal screening workflow, not a validated defect-diagnosis or plant-health assessment method, and applies only where acquisition is controlled and the offsets are recalibrated for the target camera and palette. Full article

Infrared and visible image fusion aims to integrate the complementary information from heterogeneous sensors, thereby enhancing the robustness of visual perception in complex environments. Most existing methods, however, employ fixed network parameters, a characteristic that limits their adaptive modeling capabilities for cross-modal information under scenarios such as drastic illumination changes, low light conditions, dense fog, and strong glare. To address this issue, we propose a scene-adaptive image fusion network, termed HL-Fuse, based on dynamic low-rank modulation and frequency-domain collaboration. For the spatial domain, the Hyper-LoRA is introduced via our designed SceneHyperNet, which mathematically constrains parameter variations within a low-rank subspace to adaptively calibrate attention mappings according to the global scene information. For the frequency domain, a tailored FAM is introduced to bridge spatial-domain feature aggregation and explicit spectrum reweighting by implementing targeted high- and low-frequency filtering, thereby enhancing edge and texture representation. Experiments conducted on the MSRS, TNO, M³FD, and FMB datasets demonstrate that HL-Fuse achieves competitive performance in terms of both multiple objective metrics and subjective visual quality, while the overall performance in the MSRS downstream object detection task is also enhanced. These results indicate the potential value of HL-Fuse for complex scene perception and remote sensing applications. Full article

Accurate mapping of crop residue cover (CRC) is important for sustainable agricultural management because residue supports soil health, erosion control, and carbon sequestration. Traditional ground-based CRC methods are labor-intensive and spatially limited, increasing interest in UAV-based remote sensing. UAVs can carry RGB, multispectral, and hyperspectral sensors, which capture different spectral information for distinguishing crop residue from soil. This study compared four high-spatial-resolution (2.5 cm) UAV imagery types—RGB, multispectral, visible–near-infrared (VNIR) hyperspectral, and shortwave infrared (SWIR) hyperspectral—for fine-scale CRC classification. A fully connected neural network (FCNN) was developed to classify residue and soil pixels. Performance was evaluated using two complementary approaches: pixel-level accuracy assessment based on manually delineated image samples and plot-level validation against residue percentages derived from ground photos. Results showed that high pixel-level classification accuracy values were achieved across all imagery types, with overall accuracies above 94%. However, plot-level validation revealed that sensor performance depended on the evaluation metric considered. Multispectral imagery produced the highest R² with ground photo-derived reference CRC values (R² = 0.672). These results indicate that greater spectral dimensionality did not necessarily improve plot-level CRC estimation under the tested field conditions. More importantly, the findings show that high pixel-level classification accuracy does not necessarily translate into stronger plot-level CRC estimation, highlighting the importance of using complementary validation approaches when evaluating UAV-based CRC estimation. Full article

To address the susceptibility of single-modality approaches to illumination variations and imaging conditions in low-altitude unmanned aerial vehicle (UAV) detection, this study proposes a structure-aware heterogeneous dual-stream detection network based on infrared and visible-light fusion. First, a multimodal UAV detection dataset oriented toward complex low-altitude scenarios is constructed, providing a data foundation for cross-modal detection research. Then, a structure-aware heterogeneous dual-stream feature extraction framework is designed to enable collaborative modeling of visible-light and infrared features through modality-specific encoding. In the visible-light branch, a Structure-Aware Gated Enhancement Block (SAGE Block) is introduced to enhance the representation of fine-grained structural and edge information. In the cross-modal fusion stage, a Bidirectional Wavelet-Guided Fusion Module (BWFM) is proposed to decouple structural semantics and detailed information in the frequency domain. Adaptive fusion is further achieved through low-frequency cross-modal interaction and high-frequency detail-preservation strategies. Finally, the proposed method is experimentally validated on the proposed Multispectral UAV Detection Dataset (MUDD) and the Multi-scenario Multi-Modality Fusion Dataset (M3FD).. The experimental results show that the proposed method achieves an mAP@0.5 of 0.9680 and an mAP@0.5:0.95 of 0.6794 on the proposed MUDD, as well as an mAP@0.5:0.95 of 0.6072 on the M3FD dataset, demonstrating competitive detection accuracy and generalization capability. Ablation experiments further indicate that the SAGE Block, BWFM, and the low-frequency cross-modal fusion and high-frequency detail-preservation strategies within BWFM all contribute positively to performance improvement. Full article

Multicomponent beef adulteration is challenging to detect rapidly due to high concealment, complex composition, and overlapping spectral features. This study employed visible–near-infrared hyperspectral imaging (400–1000 nm) combined with a Deep Feature-Enhanced Partial Least Squares Regression (DF-PLSR) model for quantitative detection and visualization of adulteration in beef. A total of 2322 samples were prepared across seven adulteration systems—single (chicken, duck, pork), binary, and ternary—with adulteration ratios from 5% to 50%. The DF-PLSR model outperformed traditional PLSR in all systems, achieving the best performance for chicken adulteration (R²p = 0.9942, RMSEP = 1.31%), with RPD > 5 for all systems. CARS and SPA reduced spectral dimensionality by 85.9% while maintaining R²p > 0.93. Pixel-level visualization achieved prediction errors < 3%, enabling intuitive identification of adulterant spatial distribution. This method provides a rapid, non-destructive, and accurate approach for detecting complex adulteration in beef, with strong potential for food safety monitoring. Full article