Search Results (95)

Garlic, a significant global specialty economic crop, is currently facing severe challenges from labor shortages and escalating production costs. Achieving full-process mechanized production is the core approach to ensuring sustainable industrial development and enhancing international competitiveness. This paper systematically reviews the research progress and application status of mechanized equipment throughout the entire crop cycle of garlic production, including seeding, field management, harvesting, and post-harvest processing and sorting. The study reveals that garlic equipment is undergoing a profound transformation from traditional mechanization to “opto-electro-mechanical integration” and intelligence. In the seeding phase, breakthroughs have been made in pneumatic precision seed-metering and machine vision-based clove bud orientation technologies, significantly improving the quality of upright planting. In field management, precise variable-rate application and targeted weeding have been preliminary realized through plant protection Unmanned Aerial Vehicle (UAV) downwash airflow field simulation (CFD) and deep learning-based image segmentation. In the harvesting phase, relying on 3D Discrete Element Method (3D-DEM) soil-cutting simulation and adaptive profile root-trimming technology, the industry is accelerating the transition from inefficient segmented harvesting to low-damage combined harvesting. In the post-harvest phase, hyperspectral imaging (HSI) and multi-label convolutional neural networks (CNNs) have been utilized to achieve high-speed non-destructive detection of internal and external quality. However, industry still faces critical bottlenecks such as the insufficient integration of machinery and agronomy, poor robustness of intelligent perception algorithms in complex environments, and high damage rates of core soil-engaging components. Future research should focus on lightweight algorithm deployment, digital twin-driven virtual prototyping, and the construction of regional standardized machinery–agronomy systems, aiming to build an efficient and universal intelligent production closed-loop for garlic. Full article

Convolutional neural networks (CNNs) and training-free CNN variants have been successfully applied to hyperspectral image (HSI) processing and analysis. Training-free CNNs have shown promising feature extraction performance, which could effectively address the issue of typical CNNs being highly parameterized; however, inevitable noise and redundancy in the randomly selected training-free convolutional kernels often leads to unsatisfactory performance. To address this issue, we propose Superpixel Random Selection Random Walk Multi-Branch Depthwise Convolutional Neural Network (SRSRWMD-CNN). Specifically, we propose a novel training-free convolutional neural network characterized by inter-layer multi-scale integration and intra-layer grouping. Various superpixels groups are first generated through multi-scale superpixel segmentation algorithms, then the predetermined number of superpixels are randomly sampled from these groups to serve as training-free convolution kernels. This mechanism enables adaptive computation of HSI feature maps without costly model training in the feature extraction stage, allowing the network to effectively capture a multi-scale spectral–spatial feature representation. Additionally, we propose a multi-branch depthwise convolution strategy that mitigates feature learning errors while significantly enhancing feature representation capabilities. A random walk strategy is employed to expand the receptive field and enhance the robustness of the training-free convolution kernels. Finally, the multi-scale spectral–spatial features are concatenated with the multiple convolutional stages to fuse salient shallow and deep features for accurate HSI classification. Extensive experiments demonstrate that the proposed method achieves superior performance compared to state-of-the-art algorithms. Full article

Historic stone-built heritage is continually exposed to environmental stressors that promote material degradation and surface alteration, often in spatially heterogeneous ways. Rapid, non-destructive diagnostic tools capable of capturing both spectral and spatial information are therefore essential to support preventive conservation strategies. In this study, short-wave infrared hyperspectral imaging (SWIR-HSI), combined with chemometric analysis, three-dimensional (3D) visualisation, and complementary spectroscopic techniques, is investigated as an integrated framework for assessing the conservation state of historical stonework. A field campaign was conducted at the 15th- to 17th-century San Emeterio and San Celedonio Church (Larrabetzu, Spain), a sandstone structure exposed to environmental pollution and adverse conditions. SWIR hyperspectral images (1000–2500 nm) were acquired in situ and analysed using Principal Component Analysis (PCA) and K-Means clustering to explore spectral variability and segment the façade into spectrally homogeneous regions. The resulting chemometric outputs were projected onto a photogrammetry-based 3D RGB model, enabling volumetric visualisation of material heterogeneity and surface alteration patterns. To support the interpretation of hyperspectral features, selected regions were further analysed using X-ray fluorescence (XRF) and Raman spectroscopy. The proposed 3D-SWIR approach enhances the interpretability of hyperspectral data by embedding it within its architectural context and linking spectral variability to underlying physicochemical processes. This integrated methodology demonstrates strong potential as a non-destructive diagnostic and decision-support tool for assessing, monitoring, and conserving cultural heritage stone structures. Full article

Efficient phenotyping is essential for accelerating genetic improvement in turfgrass breeding, where manual measurements are labor-intensive. This study evaluated hyperspectral imaging (HSI) as a high-throughput tool for assessing Zoysia spp. breeding populations consisting of 464 genotypes. HSI data (400–1000 nm) were processed through a user-in-the-loop hybrid segmentation pipeline integrating UMAP dimensionality reduction, DBSCAN clustering, Random Forest classification, and pseudo-RGB refinement. To independently assess vegetation classification performance, 10,000 manually annotated reference points from 50 pseudo-RGB images were compared with the automated module, yielding an overall accuracy of 0.9697, a precision of 0.8830, a recall of 0.9240, a specificity of 0.9779, an F1-score of 0.9030, and Cohen’s kappa of 0.8851. A Combined Ranking Score (CRS) integrating five vegetation indices and vegetation pixel count was significantly associated with aerial shoot count (r = −0.445, p < 0.001) and runner count (r = −0.207, p < 0.001). The highest-ranked genotype showed a 9370.3-pixel increase in vegetation area between 6 and 16 weeks after transplanting, compared with 1417.7 pixels for the lowest-ranked genotype. Classification performance declined under high-coverage conditions, indicating increased mixed-pixel ambiguity in dense canopies. These results suggest that HSI-based CRS can support rapid, objective, and non-destructive relative ranking of density-related vegetative growth in turfgrass breeding. Because the study was conducted at a single location and season and correlations with manual traits were moderate, the framework is best interpreted as a screening and ranking tool rather than a direct predictive model. Full article

Background: Accurate in vivo brain tumor detection using hyperspectral imaging (HSI), a non-invasive technique that captures spectral information beyond the visible range, is challenging due to the complexity of biological tissues and the difficulty in distinguishing malignant from healthy areas. Conventional neural-network-based methods often misclassify tumor tissue as blood vessels, largely due to high vascularization and the scarcity of annotated data. Method: To address this issue, this work proposes an underexplored approach that decomposes the problem into two tasks: (1) segmentation of the brain cortical surface and its blood vessels, and (2) segmentation of biological tissues within the segmented craniotomy site. The cortical segmentation task is addressed independently of the segmentation model used in the second stage. To achieve this, a set of pseudo-labels is generated from RGB and HSI captures acquired during in vivo brain surgeries. These pseudo-labels support a multimodal training strategy that leverages both imaging domains, yielding a model capable of segmenting the craniotomy site and the blood vessels contained in it. The model is further refined on HSI using weakly supervised fine-tuning with sparse ground truth annotations. Results: The final segmentation map combines cortical and tissue segmentation outputs, considering only cortex pixels not overlapped by vessels as potential tumor regions. This simplifies the HSI tissue segmentation task, reframing it as a binary segmentation of healthy vs. other tissues, while still enabling a comprehensive multiclass output. Conclusions: The proposed method achieves up to a 15.48% increase in F1 score for the tumor class, while segmenting the brain cortex with a mean Dice similarity coefficient (DSC) of 92.08% and accurately detecting 95.42% of labeled blood vessel samples in the HSI dataset. Full article

Hyperspectral imaging (HSI) extends conventional histopathology by combining spatial morphology with rich spectral information that reflects tissue biochemical composition, offering new opportunities for quantitative tissue analysis. However, reliable spectral analysis requires accurate instance-level segmentation of cell nuclei to enable the construction of meaningful nuclear spectral databases. In this work, a comprehensive methodology for generating hyperspectral databases of cell nuclei from histopathological samples is presented, including hyperspectral acquisition, preprocessing, nucleus segmentation, and spectral signature extraction. Three nucleus segmentation methods are evaluated: a spectral-only approach based on pixel-wise hyperspectral signatures in the visible–VNIR range; a spatial-only approach using synthetic RGB images derived from hyperspectral cubes; and a combined spatial–spectral approach that jointly exploits spatial and spectral information. The methods are assessed on a proprietary dataset of 30 hyperspectral cubes of tumor and healthy histopathological brain tissue annotated by expert pathologists. The spectral-only method achieves a Dice similarity coefficient (DSC) of 61.89% and produces severe over-segmentation, with cell count deviations exceeding substantially the ground truth in healthy tissue. The spatial-only method attains the highest pixel-wise accuracy (78.97% DSC) but underestimates nucleus counts by approximately 30% in tumor regions due to nucleus merging. The spatial–spectral method achieves a DSC of 73.13% and a mean cell count deviation of 4%, providing more reliable instance-level separation. These findings demonstrate that pixel-wise accuracy alone is insufficient for hyperspectral nuclear database generation. Full article

Accurate estimation of hematoma age remains a major challenge in forensic practice, as current assessments rely heavily on subjective visual interpretation. Hyperspectral imaging (HSI) captures rich spectral signatures that may reflect the biochemical evolution of hematomas over time. This study evaluates whether a convolutional neural network (CNN) integrating both spectral and spatial information improves hematoma age estimation accuracy. Additionally, we investigate whether performance can be maintained using a reduced, physiologically motivated subset of wavelengths. Using a dataset of forearm hematomas from 25 participants, we applied radiometric normalization and SAM-based segmentation to extract $64\times 64\times 204$ hyperspectral patches. In leave-one-subject-out cross-validation, the CNN outperformed a spectral-only Lasso baseline, reducing the mean absolute error (MAE) from 3.24 days to 2.29 days. Band-importance analysis combining SmoothGrad and occlusion sensitivity identified 20 highly informative wavelengths; using only these bands matched or exceeded the accuracy of the full 204-band model across early, middle, and late hematoma stages. These results demonstrate that spectral–spatial modeling and physiologically grounded band selection can enhance estimation accuracy while significantly reducing data dimensionality. This approach supports the development of compact multispectral systems for objective clinical and forensic evaluation. Full article

Soluble solid content (SSC) is a key indicator of strawberry quality. Conventional SSC measurement methods are destructive and impractical for large-scale applications. Therefore, this study developed a region-based hyperspectral imaging (HSI) and lightweight one-dimensional convolutional neural network (1D CNN) framework for nondestructive SSC prediction in strawberries. To evaluate spatial effects on predictive accuracy, the fruit surface was segmented into five groups (G1–G5). Three spectral preprocessing methods were applied with partial least squares regression and five convolutional neural network (CNN) architectures, including a simplified VGG-CNN. Larger regions generally improved prediction performance; however, the 50% region (G2) and 75% region (G3) achieved comparable performance to the full region, reducing data requirements. The simplified VGG-CNN model with SNV outperformed other models, exhibiting high accuracy with reduced computational cost, supporting its potential integration into portable and real-time sensing systems. The proposed approach can contribute to improved postharvest quality control and enhanced consumer confidence in strawberry products. Full article

Crop classification of farmland is of great significance for crop monitoring and yield estimation. Airborne hyperspectral systems can provide large-format hyperspectral farmland images. However, traditional machine learning-based classification methods rely heavily on handcrafted feature design, resulting in limited representation capability and poor computational efficiency when processing large-format data. Meanwhile, mainstream deep-learning-based hyperspectral image (HSI) classification methods primarily rely on patch-based input methods, where a label is assigned to each patch, limiting the full utilization of hyperspectral datasets in agricultural applications. In contrast, this paper focuses on the semantic segmentation task in the field of computer vision and proposes a novel HSI crop classification framework named All-Attention Transformer (A²Former), which combines CNN and Transformer based on a fully attention-based mechanism. First, a CNN-based encoder consisting of two blocks, the overlap-downsample and the spectral–spatial attention weights block (SSWB) is constructed to extract multi-scale spectral–spatial features effectively. Second, we propose a lightweight C-VIT block to enhance high-dimensional features while reducing parameter count and computational cost. Third, a Transformer-based decoder block with gated-style weighted fusion and interaction attention (WIAB), along with a fused segmentation head (FH), is developed to precisely model global and local features and align semantic information across multi-scale features, thereby enabling accurate segmentation. Finally, a checkerboard-style sampling strategy is proposed to avoid information leakage and ensure the objectivity and accuracy of model performance evaluation. Experimental results on two public HSI datasets demonstrate the accuracy and efficiency of the proposed A²Former framework, outperforming several well-known patch-free and patch-based methods on two public HSI datasets. Full article

Magnetic resonance imaging (MRI) and hyperspectral imaging (HSI) provide complementary information for image-guided neurosurgery, combining high-resolution anatomical detail with tissue-specific optical characterization. This work presents a novel multimodal phantom dataset specifically designed for MRI–HSI integration. The phantoms reproduce a three-layer tissue structure comprising white matter, gray matter, tumor, and superficial blood vessels, using agar-based compositions that mimic MRI contrasts of the rat brain while providing consistent hyperspectral signatures. The dataset includes two designs of phantoms with MRI, HSI, RGB-D, and tracking acquisitions, along with pixel-wise labels and corresponding 3D models, comprising 13 phantoms in total. The dataset facilitates the evaluation of registration, segmentation, and classification algorithms, as well as depth estimation, multimodal fusion, and tracking-to-camera calibration procedures. By providing reproducible, labeled multimodal data, these phantoms reduce the need for animal experiments in preclinical imaging research and serve as a versatile benchmark for MRI–HSI integration and other multimodal imaging studies. Full article

Background: Scabies, caused by Sarcoptes scabiei var. hominis, remains difficult to confirm histologically when parasites are sparse or fragmented. Conventional microscopy is particular but limited by small sample size, tissue destruction, and observer dependence. Objective: To evaluate visible–near-infrared hyperspectral imaging (VIS–NIR HSI) as a label-free optical method for detecting S. scabiei in human skin sections and to assess its compatibility with routine HE staining. Methods: Formalin-fixed, paraffin-embedded (FFPE) skin tissue from six patients with histologically verified scabies was analysed using VIS–NIR HSI (500–1000 nm). Unstained sections mounted on CaF₂ substrates and parallel HE-stained slides were imaged. Spectral datasets were processed by principal component analysis and segmentation to distinguish mite structures from epidermal and dermal compartments. Results: The chitin-rich mite exoskeleton exhibited a reproducible reflectance slope in the near-infrared range (R₈₅₀/R₅₅₀ > 1.5), clearly separating parasite from host tissue (R₈₅₀/R₅₅₀ < 1.0). PCA confirmed consistent cluster separation across all cases (ΔPC ≈ 3.7 ± 0.2). These contrasts remained detectable in HE-stained sections, validating applicability to conventional slides. Conclusions: VIS–NIR HSI enables reliable, label-free detection of S. scabiei mites in both unstained and HE-stained human skin tissue. By combining morphological and biochemical information in a single modality, HSI represents a promising adjunct to digital dermatopathology and may improve diagnostic sensitivity in challenging or atypical cases. Full article

Hyperspectral image (HSI) target detection plays a critical role in both military and civilian applications, including military reconnaissance, environmental monitoring, and precision agriculture. However, the complex background of the scene severely restricts the further improvement of hyperspectral target detection performance. To address this challenge, we propose a diffusion model hyperspectral target detection method based on multivariate Gaussian background noise. The method constructs multivariate Gaussian-distributed background noise samples and introduces them into the forward diffusion process of the diffusion model. Subsequently, the denoising network is trained, the conditional probability distribution is parameterised, and a designed loss function is used to optimise the denoising performance and achieve effective suppression of the background, thus improving the detection performance. Moreover, in order to obtain accurate background noise, we propose a background noise extraction strategy based on spatial–spectral centre weighting. This strategy combines with the superpixel segmentation technique to effectively fuse the local spatial neighbourhood information of HSI. Experiments conducted on four publicly available HSI datasets demonstrate that the proposed method achieves state-of-the-art background suppression and competitive detection performance. The evaluation using ROC curves and AUC-family metrics demonstrates the effectiveness of the proposed background-suppression-guided diffusion framework. Full article

Mold contamination is one of the critical factors threatening the safety of paper-based cultural relics. Current detection methods rely predominantly on offline analysis, facing challenges such as low efficiency and limited real-time accuracy, which hinder their effectiveness in meeting the technical requirements of cultural heritage preventive conservation. This study proposes a hyperspectral imaging (HSI)-deep learning integrated fungal segmentation framework for deterioration detection in paper-based artifacts. Firstly, the HSI data was reduced to three dimensions via Locally Linear Embedding (LLE) manifold learning to construct 3D pseudo-color imagery, effectively preserving discriminative spectral features between fungal colonies and substrates while eliminating spectral redundancy. Secondly, a hybrid architecture synergizing Feature Pyramid Networks (FPN) with Vision Transformers was developed for semantic segmentation, leveraging CNN’s local feature extraction and Transformer’s global context modeling to enhance fungal signature saliency and suppress background interference. Innovatively, a dynamic sparse attention mechanism is introduced, optimizing attention allocation through the TOP-K algorithm to screen regions richer in mold information spatially and spectrally, thereby improving segmentation accuracy. Semantic segmentation experiments were conducted on papers infected with different molds. The results demonstrate that the proposed method achieves excellent performance in mold segmentation, providing technical support for mold detection and preventive conservation of cultural relics. Full article

Ensuring food safety and quality is critical in the food-processing industry, where the detection of contaminants remains a persistent challenge. This study assesses the feasibility of hyperspectral imaging (HSI) for detecting foreign objects on pork belly meat. A Specim FX17 hyperspectral camera was used to capture data across various bands in the near-infrared spectrum (900–1700 nm), enabling identification of contaminants that are often missed by traditional visual inspection methods. The proposed solution combines a segmentation approach based on a lightweight Vision Transformer with specific pre- and post-processing strategies to distinguish contaminants from meat, fat, and conveyor belt, while emphasizing on a low false-positive rate. On a test set of 55 images with contaminants, the method retained most true positives; on 183 clean images, the full pipeline eliminated all false positives. Across 208 additional images acquired under production-line temperature variation (10–55 °C), only one image exhibited small false positives, and on a challenging 95-image set with fat-like spectra the pipeline produced zero false positives. These results demonstrate high detection accuracy and training efficiency while addressing issues such as noise, temperature drift, and spectral similarity. The findings support the feasibility of real-time HSI for automated quality control. Full article

Effective feature extraction is a key issue in hyperspectral image (HSI) classification task. Recent works have studied hyperspectral classification models based on various deep architectures. However, the specific architecture cannot fully exploit the complementary diversity of global and local features in HSIs, resulting in suboptimal results. To address these issues, we fully utilize the advantages of GNN and CNN in global and local feature extraction and design a new end-to-end global–local feature fusion enhancement network (GLFFEN). Specifically, we first construct a GNN with dynamically weighted neighbor contributions using superpixel-segmented patches as nodes, named the Graph Attention (GA) branch. Additionally, we design a spatial–spectral feature attention module (SSFAM) to enhance the ability of the CNN to extract spatial and spectral features in local neighborhoods, termed the spatial–spectral feature attention (SSFA) branch. Moreover, a multi-feature adaptive fusion (MAF) module is proposed to solve the problem of weight distribution during global–local feature fusion. Experiments on three well-known HSI datasets have shown that our GLFFEN surpasses state-of-the-art (SOTA) methods on three widely used metrics. Full article