Search Results (265)

Infrared target recognition plays an important role in many real-world applications, but its performance is often constrained by the scarcity of annotated infrared data. To alleviate this issue, optical-to-infrared image translation has been widely explored as a data augmentation strategy by leveraging the abundance of optical images. However, existing approaches typically overlook the intrinsically multimodal nature of optical-to-infrared mapping, leading to insufficient diversity in the synthesized infrared images. Moreover, the lack of effective constraints to preserve semantic fidelity further hampers the practical utility of generated samples for recognition tasks. In this paper, we propose a multimodal style translation framework for infrared target recognition. The proposed framework is built upon a style-semantic disentanglement architecture, which decouples domain-general semantic structures from domain-specific style statistics, thereby enabling flexible recombination of optical content with diverse infrared characteristics. Furthermore, we design a multi-level adaptive loss function that explicitly enforces complementary constraints on structural fidelity and semantic consistency during the translation process. Extensive experiments on two public datasets demonstrate the effectiveness of SSD-VI. On RGB-NIR, it achieves an FID of 46.53 and a KID of 0.0331, while increasing classification accuracy by 6.68 percentage points, from 83.37% to 90.05%. On VEDAI, SSD-VI improves mAP@50 by 0.13 for YOLOv8m and 0.14 for RT-DETR, confirming the value of the generated samples for infrared target recognition. Full article

Weed segmentation in soybean fields is essential to improving herbicide use efficiency and supporting precision variable-rate spraying. This study developed an SiLU–CPCA–Gate U-Net (SCG-UNet) using fused UAV RGB and multispectral imagery to improve the delineation of small and partially occluded weeds under complex canopy conditions. SCG-UNet integrates channel–spatial feature enhancement, attention-guided skip-feature fusion, and smooth nonlinear activation within a U-Net framework. A total of 400 spatially aligned RGB–multispectral image groups collected from a soybean field in Henan Province, China, were manually annotated for model development and evaluation. Paired bootstrap comparisons showed that RGB+NIR achieved the highest numerical performance among the tested inputs and significantly outperformed RGB, RGB+R, and RGB+G in mIoU after Holm correction, while remaining statistically comparable to RGB+REdge and RGB+NIR+REdge. With RGB+NIR input, SCG-UNet achieved an mPA of 92.35%, an mIoU of 83.43%, a Dice coefficient of 79.50%, and an F1-score of 80.77%, exceeding the baseline U-Net by 0.71, 1.50, 2.19, and 2.09 percentage points, respectively. Five-fold spatial block cross-validation yielded an mIoU of 82.92 ± 0.29% and an F1-score of 80.06 ± 0.40%, indicating stable performance across different regions of the same field. SCG-UNet also achieved the highest numerical mIoU among the evaluated convolutional, high-resolution, and Transformer-based models, exceeding TransUNet and LeViT-UNet by 0.90 and 0.71 percentage points, respectively, while requiring fewer parameters and lower reported memory consumption. The segmentation results were further converted into a conceptual variable-rate spraying prescription map with five spray volume levels ranging from 220 to 300 L/ha. These results demonstrate the potential of RGB–multispectral fusion for soybean weed mapping, although field validation of prescription execution, weed control efficacy, and economic benefits remains necessary. Full article

Copper-deficient copper selenide (Cu_2−xSe) nanocrystals possess strong near-infrared (NIR) absorption and efficient photothermal conversion, making them attractive candidates for photoacoustic imaging. In this study, Cu_2−xSe nanocrystals with distinct morphologies were synthesized using different selenium precursors and evaluated as photoacoustic contrast agents. Se–oleylamine precursors produced predominantly disk-shaped nanocrystals with average dimensions of approximately 20 nm in diameter and 5 nm in thickness, while Se–TOP/TOPO precursors yielded smaller spherical nanocrystals. Structural characterization by transmission electron microscopy, high-resolution TEM, and selected-area electron diffraction confirmed the formation of highly crystalline copper-deficient Cu_2−xSe nanocrystals with a face-centered cubic crystal structure. UV–Vis–NIR spectroscopy revealed broad optical absorption extending into the NIR region, with morphology-dependent spectral characteristics. Multispectral optoacoustic tomography demonstrated strong photoacoustic signal generation from both nanodisks and nanospheres over a broad wavelength range. In vivo studies using PEGylated Cu_2−xSe nanospheres showed successful lymphatic uptake following hind paw injection and enabled visualization of the draining popliteal lymph node through spectral unmixing of nanoparticle and hemoglobin signals. These results demonstrate that Cu_2−xSe nanocrystals are promising photoacoustic contrast agents for lymphatic imaging and other biomedical imaging applications. Full article

Quality deterioration of chilled lamb during storage poses a challenge to meat quality and safety control, making rapid and accurate freshness-grade classification essential. Existing methods based on either spectral information or RGB image information alone are insufficient to simultaneously characterize internal chemical changes and external appearance changes during lamb quality deterioration. To address this issue, this study developed a chilled lamb freshness-grade classification method by integrating deep features from two-dimensional visible–near-infrared (Vis–NIR) spectral encoding with RGB image features. In this method, one-dimensional Vis–NIR spectra were transformed into two-dimensional encoded images using Gramian angular difference field (GADF), Gramian angular summation field (GASF), Markov transition field (MTF), and recurrence plot (RP) to enhance the representation of inter-wavelength structural relationships in spectral sequences, thereby compensating for the limited ability of conventional one-dimensional spectral modeling to capture global correlations and local variation information. Meanwhile, recursive feature elimination (RFE)-selected spectral deep features were fused with Spearman-selected RGB image features to construct a deep–shallow classification model. The results showed that the fusion models outperformed the single-modality models, with GADF(10%)+Image-SVM achieving the best performance, yielding an accuracy, F1-score, and MCC of 0.966, 0.957, and 0.946, respectively. Shapley additive explanations (SHAP) analysis further indicated that GADF deep features were the primary contributors, while RGB image features provided effective complementary information, demonstrating the potential of the proposed method for rapid and nondestructive freshness-grade classification of chilled lamb. Full article

The twelfth-century Beneventan manuscripts of the Cathedral Chapter Library of Benevento constitute a largely understudied corpus of medieval illumination, whose material composition has not yet been investigated. This paper presents an in situ non-invasive multi-analytical investigation of three of these codices: Ms. 7 and Ms. 17 (Sanctorale), and Ms. 19 (Missal–Breviary) assigned to two different production environments, through a workflow combining elemental/molecular spectroscopy with point-based and Vis-NIR hyperspectral imaging and high-resolution microscopy. The study examines pigments, binders, and inks, across the three manuscripts, with the aim of establishing an exploratory research protocol that combines autoptic observations and diagnostic analyses to characterising and comparing manuscript production environments. The principal pigments identified across the three manuscripts are cinnabar, natural ultramarine (lapis lazuli), a copper-based green, and lead white, with iron-gall ink consistently present for text writing. Notable differences between the two production environments include the use of orpiment versus an unidentified organic lake for yellow, and the presence of minium, a red–purple lake, and gold exclusively in Ms. 19. A particularly significant finding is the use of iron gall ink, employed across all three manuscripts as a writing medium and as a pigment for modulating tonal values. Moreover, a brown ink was identified in preparatory underdrawings and decorative details, but its composition remains undetermined. Full article

Rapid and non-destructive authentication of herbal medicines is important for quality control and market supervision. This study established an interpretable spectral evidence learning framework for visible and near-infrared (Vis/NIR) imaging-based authentication of Codonopsis Radix (CR) and Aurantii Fructus (AF). Compact 31-band mean gray-value spectra were analyzed at ROI and sample levels. CR sample-level spectra were obtained by ROI-group averaging, whereas AF records were retained as individual sample spectra with image-group information used for leakage-controlled validation. Raw spectra, Savitzky–Golay smoothing, multiplicative scatter correction, and standard normal variate correction were compared with machine-learning and deep-learning classifiers. A fold-contained lightweight diffusion (LD) module was further introduced to provide class-conditioned spectral augmentation and denoising-error evidence. Under grouped cross-validation, the strongest non-LD Linear SVM models achieved accuracy/macro-F1 values of 0.9231/0.9238 for CR and 0.9025/0.9018 for AF. After LD augmentation, the best LD-augmented SVM models reached macro-F1 values of 0.9427 and 0.9197, respectively. Across all evaluated model–dataset combinations, LD increased the overall mean macro-F1 from 0.7302 to 0.8189. Model-aligned wavelength evidence and top-wavelength subset tests further showed that selected LED-band subsets retained useful discriminative information within the present imaging configuration. These results support the feasibility of compact Vis/NIR image-based authentication of herbal materials under grouped validation. Full article

As hen eggs are a primary source of high-quality dietary protein, egg freshness is fundamentally linked to biochemical alterations during storage, including moisture redistribution, protein degradation, and fluctuating chemical profiles. Accurate assessment of these internal changes is paramount for quality control; nonetheless, conventional analytical techniques remain predominantly destructive, rendering them impractical for high-throughput industrial monitoring. While existing literature has explored individual spectroscopic methods, the synergistic potential of multi-sensor integration and advanced artificial intelligence (AI) algorithms remains insufficiently synthesized. This review systematically evaluates recent breakthroughs in integrating AI with diverse spectroscopic modalities for non-destructive freshness quantification, including Visible-Near-Infrared (VIS-NIR), Raman, Fluorescence, and Hyperspectral Imaging (HSI). We elucidate the underlying mechanisms of spectral response to internal quality degradation and discuss the evolution of data-driven modeling from traditional chemometrics to sophisticated deep learning architectures. Furthermore, this work identifies critical bottlenecks in real-time industrial implementation and proposes future research trajectories toward intelligent multi-sensor fusion platforms. Full article

Moldy pear core constitutes a severe internal defect that compromises fruit quality. This study proposes a nondestructive detection method for Korla pear moldy core using Vis/NIR spectral signals, aimed at supporting post-harvest quality control and automated industrial sorting. We collected spectral signals from pears and quantified the moldy pear core area to classify samples into healthy (S = 0%), slightly moldy (0 < S ≤ 10%), and severely moldy (S > 10%) categories. We constructed a three-tier comparative framework to evaluate the progression from conventional machine learning to advanced deep learning: traditional methods using univariate selection (US) and random forest (RF) for feature extraction followed by support vector machine (SVM) classification; 1D-ResNet for direct processing of spectral signals; and two-dimensional approaches transforming signals into improved gramian angular field (IGAF) or Laplacian pyramid Markov transition field (LPMTF) images processed through deep belief network (DBN), MobileNetv3, and Vision Transformer (ViT). The LPMTF-ViT combination delivered the best performance with 98.98% test accuracy and 94.44% external validation accuracy, significantly exceeding traditional approaches and 1D-ResNet. This innovative approach delivers effective technical support for early-stage, nondestructive detection of internal fruit defects. It also establishes a scalable foundation for automated industrial inspection systems, potentially reducing post-harvest losses while ensuring premium quality control in modern fruit supply chains. Full article

This study addresses the early detection of water stress in grapevines (Vitis vinifera L. cv. Monastrell), a key challenge for precision irrigation. The main objective is to assess the feasibility of VIS–NIR hyperspectral imaging (400–1000 nm) to anticipate water stress, relating the spectral signal to stem water potential. This study was developed over two campaigns, in 2024 and 2025, using 18 potted plants. In 2024, eight vines were irrigated, and the remaining 10 were subjected to water-deprivation treatments, whilst in 2025, all plants were irrigated, but half at a control dose and the rest at a reduced dose equivalent to 33% of the control. Images were acquired over five dates in June 2024 and over seven in June 2025 using a Specim IQ camera; stem potential was also measured to provide a physiological reference. Individual time series were developed, calculating the Mahalanoubis distance in a PCA space. Results revealed a change window between 10 and 13 June, consistent with the divergence in water potential from 17 to 24 June. PCA highlighted spectral regions related to changes in pigments, nitrogen and water content as main indicators of water stress. We conclude that HSI is a promising tool for early water stress detection. Full article

Remote sensing is widely used in archaeological prospection to detect surface anomalies (crop marks) indicating buried remains, typically through recognition of visual patterns in high- or very high-resolution imagery acquired by means of satellite, airborne, or drone sensors. In contrast, spectroscopic approaches focusing on variations in spectral signatures still remain rarely applied in archaeological research. This study proposes a technological barrier-free method addressed to archaeologists which is based on pixel-level analysis of the Reflectance Values (RV) and spectral shape variations in the visible, near-infrared and short-wave infrared (VIS-NIR-SWIR) range derived from Sentinel-2 imagery. Spectral signatures are extracted through sampling polygons designed to account for the spatial resolution of the different Sentinel-2 bands and their spatial relationship with the location and size of the archaeological features. The RV method is tested on two Roman archaeological contexts: the ancient city of Telesia Vetere (San Salvatore Telesino, Benevento) and a Roman villa at Podere Colle Agnano (Labro, Rieti) using the full Sentinel-2 archive since 2017. While Telesia has previously been investigated through aerial photo interpretation and archaeological fieldwork, the Roman villa at Labro is documented here for the first time. Results show consistent seasonal repeated spectral separability between areas corresponding to known buried archaeological features and surrounding areas. Similar anomalies were also detected in areas without previously documented remains, thus suggesting the possible presence of buried structures and highlighting the predictive potential of the RV method. Owing to its easiness to use beyond image processing specialism and reliance on open-access data, the method can support archaeological decision-making and guide further investigation with higher-resolution remote sensing data or targeted field surveys, particularly in the framework of preventive archaeology. Full article

As one of the core technologies in modern national defense and security fields, infrared stealth technology aims to realize the controllable regulation of the radiation characteristics of targets in the infrared band. This paper focuses on a novel electrochromic device with a structure of WO₃/nickel mesh/Al³⁺-Zn²⁺gel electrolyte/zinc foil. The structural composition and working mechanism are systematically analyzed, and the infrared stealth regulation performance is emphatically studied. The WO₃ thin film and device structure were characterized by scanning electron microscopy (SEM). The infrared emissivity modulation and optical response properties of the device were measured using an infrared thermal imager and a UV-Vis-NIR spectrophotometer. The prepared WO₃ film exhibits a dense spherical morphology, indicating excellent uniformity and compactness. After 1000 cycles, the areal capacitance of the device remains 83.7% of its initial value, demonstrating good cycling stability. Under the voltage regulation of −0.1 V to 1.1 V, the emissivity ε of the device at the typical mid-wave infrared wavelength of 4.0 μm decreases from 0.89 (−0.1 V) to 0.67 (1.1 V), with an absolute modulation amplitude Δε of 0.22. At the typical long-wave infrared wavelength of 8.7 μm, ε decreases from 0.96 (−0.1 V) to 0.69 (1.1 V), with an absolute modulation amplitude Δε of 0.29. The electrochromic switching times for coloring and bleaching are 10.1 s and 2.44 s, respectively. According to infrared thermal imaging tests, in the temperature range of 30–40 °C, the surface temperature difference ΔT between the colored state and bleached state increases from 4.3 °C to 4.6 °C. The maximum regulation amplitude reaches 4.6 °C at 40 °C. The device achieves efficient regulation of infrared emissivity through the electrochromic effect, providing a new device design strategy for infrared stealth technology. Full article

Reliable seed accession identification underpins germplasm conservation, traceability and breeding; however, conventional assays remain destructive, labour-intensive and difficult to scale. Here, visible–near-infrared–shortwave infrared (VIS–NIR–SWIR) hyperspectral imaging (HSI; 449.54–2399.17 nm; 563 bands) was used to classify 32 grain–legume accessions (n = 3200 seeds; 100 seeds per accession), comprising 30 common bean (Phaseolus vulgaris L.) landraces plus two outgroup legumes (Vigna angularis (Willd.) Ohwi & Ohashi and Cajanus cajan (L.) Huth). Each seed was represented by one ROI-averaged spectrum obtained from mean representative pixels within a standardised 10 × 10 pixel window at the centre of each seed. A fixed stratified 70:30 seed-level training:test partition was used, with 70 seeds per accession (n = 2240) reserved for fully independent training and 30 seeds per accession (n = 960) reserved as a fully independent test set. Principal component analysis (PCA) captured 97.42% of the spectral variance in the first three components (PC1 = 63.34%, PC2 = 23.78%, and PC3 = 10.31%). One-versus-rest wavelength association mapping revealed a maximum R² of 0.775 at 461.37 nm, and ReliefF concentrated the strongest reduced-band signal within 449.54–456.30 nm and 577.02–597.54 nm. In the original ReliefF-selected 16-band benchmark, the subspace discriminant reached 68.25% macro-F1 and 68.54% balanced accuracy; after edge-band trimming, the alternative 16-band configuration decreased to 60.67% and 60.94%, respectively. With respect to the full-spectrum sensitivity benchmark, linear discriminant analysis achieved 96.35% balanced accuracy, followed by linear SVM (94.17%). Deep learning trained directly on the full 563-band spectra reached 84.90% test accuracy, 84.47% macro-F1, 86.27% precision and 84.90% recall, with MLP_Wide outperforming the convolutional, recurrent and attention-based alternatives. Overall, under controlled laboratory conditions, this benchmark shows that accession discrimination is driven mainly by visible-domain contrasts in the most compact representations, whereas the full spectral context remains important for the most confusable accessions and for cautious future sensor design. The reduced-band findings should therefore be interpreted as exploratory guidance for sensor design rather than as a validated deployment-ready specification. Full article

Rapid, standardised discrimination of soil taxonomic units remains challenging when relying solely on conventional field descriptions and laboratory analyses, particularly at high sampling densities. This study evaluated whether proximal spectroscopy and hyperspectral imaging can support the classification of Brazilian Soil Classification System (SiBCS) suborders and pedogenetic horizons when surface and subsurface spectra are treated separately. Six intact soil monoliths (0.12 × 1.60 m) were collected in Paraná State, southern Brazil, representing one Organossolo (Ooy), three Latossolos (LVd, LVd1, and LVd2) and two Argissolos (PVAd and PVd). For each monolith, 800 spectra were acquired per sensor with a non-imaging VIS–NIR–SWIR spectroradiometer (350–2500 nm), and 800 spectra per sensor per monolith were extracted from the SWIR hyperspectral images (1200–2450 nm). Principal component analysis (PCA) was used to summarise spectral variability, and supervised classification was performed via k-nearest neighbours, random forest, decision tree and gradient boosting for suborders (10-fold cross-validation), and a neural network was used for within-profile horizon classification. PCA indicated that most of the spectral variance was captured by a dominant axis, with clearer separation among suborders in the SWIR space than in the full VIS–NIR–SWIR range. With respect to suborder classification, subsurface spectra outperformed surface spectra, and SWIR outperformed VIS–NIR–SWIR: the best accuracies were 0.96 for subsurface SWIR (gradient boosting; AUC = 0.99; MCC = 0.95) and 0.89 for surface SWIR (k-nearest neighbours; AUC = 0.98; MCC = 0.87). Within-profile horizon classification via VIS–NIR–SWIR achieved accuracies of 0.84–0.97 with the Neural Network, with most misclassifications occurring between adjacent horizons. Overall, subsurface SWIR information provided the most reliable basis for taxonomic discrimination, whereas horizon classification was feasible but reflected gradual spectral transitions along the profile. Full article

Advanced sensing technologies increasingly support monitoring and decision-making processes in modern agriculture. This study investigates the feasibility of developing a harvest timing monitoring workflow based on a portable hyperspectral imaging (HSI) system in the visible–near-infrared (VIS-NIR: 400–1000 nm) range, coupled with machine learning. A hierarchical Partial Least Squares–Discriminant Analysis (Hi-PLS-DA) model was developed and tested to discriminate harvestable from non-harvestable plants of Brassica rapa subsp. sylvestris through the identification of open flowers within otherwise closed flower buds in the raceme. The classification included four target plant classes, i.e., green inflorescences, green leaves, yellow flowers, and yellow leaves, along with two non-target classes, background and not-classified (NC), which were included to support the classification process. The predicted hyperspectral images demonstrated a clear distinction between closed and open flowers, supported by satisfactory classification performance (sensitivity, specificity, precision, and F1-score: 0.78–1.00). This workflow proved effective in handling intrinsic outdoor hyperspectral variability, mitigating illumination and canopy texture, and offers useful methodological insights for the possible future integration of HSI-based approaches into automated field applications, paving the way for rapid, real-time harvest decision support. Full article

Soybean cyst nematode (SCN) is the most economically devastating pathogen of soybean in North America. Even at low to moderate infestation levels, SCN can cause 20–30% yield loss without producing any visible aboveground symptoms. In severely infested fields, yield reductions can reach 60–70% and, in extreme cases, exceed 80%. Prior research on identifying SCN infestations has primarily relied on traditional machine-learning methods applied to Unmanned Aerial System (UAS)-based multispectral imagery, with limited success. This study hypothesizes that deep-learning (DL) methods can more effectively capture the subtle spectral and spatial signatures in multispectral images of SCN stress. To address this gap, we evaluate the performance of advanced DL architectures, including Vision Transformer (ViT) and a customized Convolutional Neural Network (CNN), for detecting SCN infestation in soybean fields using multispectral UAS imagery. Spectral analysis of the multispectral imagery revealed that the near-infrared (NIR) band is a strong discriminator between non-detected and SCN-infested areas. The DL models trained and tested across multiple growth stages showed promising results. The four-timestamp ViT model (3 June, 29 July, 19 August, and 2 September) achieved an F1-score of 0.74, while the five-timestamp SCN–CNN model (3 June, 22 July, 29 July, 19 August, and 2 September) achieved an F1-score of 0.75. Although overall performance was comparable, ViT demonstrated more stable performance across varying training and test data distributions. These findings highlight the effectiveness of DL architectures to automatically extract subtle, complex plant features from multispectral imagery throughout the growing season. Compared with manual, time-consuming soil-sampling techniques, the proposed framework enables more precise spatial and temporal monitoring of SCN infestations across fields. Full article