Recent Advances of Hyperspectral Imaging Technology and Applications in Agriculture

Microplastics or micro-bioplastics released by wrinkling paper cup

Paper cups have been widely used such as in the fast-food industry for drinking and are generally made of disposable material. To make the paper cup waterproof and prevent leakage, a thin layer of plastic such as polylactic acid (PLA) is commonly coated onto the inner wall surface. This plastic layer can potentially release debris as microplastics, particularly when the cup is wrinkled/crumpled to break and peel off the coating layer, which is tested herein. Using scanning electron microscope (SEM), the broken coating layer can be clearly observed. We then identify the coating material as plastic using mass and Raman spectra. We further employ Raman imaging to identify the released and fallen down debris as microplastics. We cross-check Raman image with SEM image to benefit each other and increase the analysis certainty, because Raman imaging can identify plastic via hyper spectrum to increase the signal-to-noise ratio, while SEM can visualise plastic with a high resolution down to micro-/nano- size. We then employ particle analysis algorithm to estimate the release amount, at approximate 180 microplastic/wrinkle, or micro-bioplastic if considering the main material of PLA as a bioplastic. Overall, we should not wrinkle the paper cup to avoid the potential release of microplastics or micro-bioplastics particularly before and during the drinking process, and the characterisation in this report might be helpful for further research on microplastics.

Attention-based Sparse and Collaborative Spectral Abundance Learning for Hyperspectral Subpixel Target Detection

The subpixel target detection in hyperspectral image processing persists as a formidable challenge. In this paper, we present a novel subpixel target detector termed attention-based sparse and collaborative spectral abundance learning for subpixel target detection in hyperspectral images. To help suppress background during subpixel target detection, the proposed method presents a pixel attention-based background sample selection method for background dictionary construction. Besides, the proposed method integrates a band attention-based spectral abundance learning model, replete with sparse and collaborative constraints, in which the band attention map can contribute to enhancing the discriminative ability of the detector in identifying targets from backgrounds. Ultimately, the detection result of the proposed detector is achieved by the learned target spectral abundance after solving the designed model using the alternating direction method of multipliers algorithm. Rigorous experiments conducted on four benchmark datasets, including one simulated and three real-world datasets, validate the effectiveness of the detector with the probability of detection of 90.88%, 96.86%, and 97.79% on the PHI, RIT Campus, and Reno Urban data, respectively, under fixed false alarm rate equal 0.01, indicating that the proposed method yields superior hyperspectral subpixel detection performance and outperforms existing methodologies.

Early diagnosis of Cladosporium fulvum in greenhouse tomato plants based on visible/near-infrared (VIS/NIR) and near-infrared (NIR) data fusion

Plant diseases can inflict varying degrees of damage on agricultural production. Therefore, identifying a rapid, non-destructive early diagnostic method is crucial for safeguarding plants. Cladosporium fulvum (C. fulvum) is one of the major diseases in tomato growth. This work presents a method of data fusion using two hyperspectral imaging systems of visible/near-infrared (VIS/NIR) and near-infrared (NIR) spectroscopy for the early diagnosis of C. fulvum in greenhouse tomatoes. First, hyperspectral images of samples at health and different times of infection were collected. The average spectral data of the image regions of interest were extracted and preprocessed for subsequent spectral datasets. Then different classification models were established for VIS/NIR and NIR data, optimized through various variable selection and data fusion methods. The principal component analysis-radial basis function neural network (PCA-RBF) model established using low-level data fusion achieved optimal results, achieving accuracies of 100% and 99.3% for calibration and prediction, respectively. Moreover, both the macro-averaged F1 (Macro-F1) values reached 1, and the geometric mean (G-mean) values reached 1 and 1, respectively. The results indicated that it was feasible to establish a PCA-RBF model by using the hyperspectral technique with low-level data fusion for the early detection of C. fulvum in greenhouse tomatoes.

Miniaturized Hyperspectral Imager Utilizing a Reconfigurable Filter Array for Both High Spatial and Spectral Resolutions

Miniaturized hyperspectral imaging based on filter arrays has attracted much attention in consumer applications, such as food safety and biomedical applications. In this Letter, we demonstrate a miniaturized hyperspectral imager using a reconfigurable filter array to tackle the existing trade-off issue between the spectral and spatial resolutions. Utilizing tens of intermediate states of a vanadium dioxide cavity, we increase the total number of physical spectral channels by tens of times from a 2 × 2 mosaic filter unit, providing both high spatial and spectral resolutions for spectral imaging. The reconfigurable filter has a good spectral resolvability of 10 nm in the visible range with a wavelength inaccuracy of less than 2.1 nm. Hyperspectral imaging is demonstrated with a frame rate of 4.5 Hz.

Performance Characterization of an Illumination-Based Low-Cost Multispectral Camera

Spectral imaging has many applications, from methane detection using satellites to disease detection on crops. However, spectral cameras remain a costly solution ranging from 10 thousand to 100 thousand euros for the hardware alone. Here, we present a low-cost multispectral camera (LC-MSC) with 64 LEDs in eight different colors and a monochrome camera with a hardware cost of 340 euros. Our prototype reproduces spectra accurately when compared to a reference spectrometer to within the spectral width of the LEDs used and the ±1σ variation over the surface of ceramic reference tiles. The mean absolute difference in reflectance is an overestimate of 0.03 for the LC-MSC as compared to a spectrometer, due to the spectral shape of the tiles. In environmental light levels of 0.5 W m-2 (bright artificial indoor lighting) our approach shows an increase in noise, but still faithfully reproduces discrete reflectance spectra over 400 nm-1000 nm. Our approach is limited in its application by LED bandwidth and availability of specific LED wavelengths. However, unlike with conventional spectral cameras, the pixel pitch of the camera itself is not limited, providing higher image resolution than typical high-end multi- and hyperspectral cameras. For sample conditions where LED illumination bands provide suitable spectral information, our LC-MSC is an interesting low-cost alternative approach to spectral imaging.

Advances in the tea plants phenotyping using hyperspectral imaging technology

Rapid detection of plant phenotypic traits is crucial for plant breeding and cultivation. Traditional measurement methods are carried out by rich-experienced agronomists, which are time-consuming and labor-intensive. However, with the increasing demand for rapid and high-throughput testing in tea plants traits, digital breeding and smart cultivation of tea plants rely heavily on precise plant phenotypic trait measurement techniques, among which hyperspectral imaging (HSI) technology stands out for its ability to provide real-time and rich-information. In this paper, we provide a comprehensive overview of the principles of hyperspectral imaging technology, the processing methods of cubic data, and relevant algorithms in tea plant phenomics, reviewing the progress of applying hyperspectral imaging technology to obtain information on tea plant phenotypes, growth conditions, and quality indicators under environmental stress. Lastly, we discuss the challenges faced by HSI technology in the detection of tea plant phenotypic traits from different perspectives, propose possible solutions, and envision the potential development prospects of HSI technology in the digital breeding and smart cultivation of tea plants. This review aims to provide theoretical and technical support for the application of HSI technology in detecting tea plant phenotypic information, further promoting the trend of developing high quality and high yield tea leaves.

Assessing the Efficacy of the Spectrum-Aided Vision Enhancer (SAVE) to Detect Acral Lentiginous Melanoma, Melanoma In Situ, Nodular Melanoma, and Superficial Spreading Melanoma

Skin cancer is the predominant form of cancer worldwide, including 75% of all cancer cases. This study aims to evaluate the effectiveness of the spectrum-aided visual enhancer (SAVE) in detecting skin cancer. This paper presents the development of a novel algorithm for snapshot hyperspectral conversion, capable of converting RGB images into hyperspectral images (HSI). The integration of band selection with HSI has facilitated the identification of a set of narrow band images (NBI) from the RGB images. This study utilizes various iterations of the You Only Look Once (YOLO) machine learning (ML) framework to assess the precision, recall, and mean average precision in the detection of skin cancer. YOLO is commonly preferred in medical diagnostics due to its real-time processing speed and accuracy, which are essential for delivering effective and efficient patient care. The precision, recall, and mean average precision (mAP) of the SAVE images show a notable enhancement in comparison to the RGB images. This work has the potential to greatly enhance the efficiency of skin cancer detection, as well as improve early detection rates and diagnostic accuracy. Consequently, it may lead to a reduction in both morbidity and mortality rates.

Combining Hyperspectral Techniques and Genome-Wide Association Studies to Predict Peanut Seed Vigor and Explore Associated Genetic Loci

Seed vigor significantly affects peanut breeding and agricultural yield by influencing seed germination and seedling growth and development. Traditional vigor testing methods are inadequate for modern high-throughput assays. Although hyperspectral technology shows potential for monitoring various crop traits, its application in predicting peanut seed vigor is still limited. This study developed and validated a method that combines hyperspectral technology with genome-wide association studies (GWAS) to achieve high-throughput detection of seed vigor and identify related functional genes. Hyperspectral phenotyping data and physiological indices from different peanut seed populations were used as input data to construct models using machine learning regression algorithms to accurately monitor changes in vigor. Model-predicted phenotypic data from 191 peanut varieties were used in GWAS, gene-based association studies, and haplotype analyses to screen for functional genes. Real-time fluorescence quantitative PCR (qPCR) was used to analyze the expression of functional genes in three high-vigor and three low-vigor germplasms. The results indicated that the random forest and support vector machine models provided effective phenotypic data. We identified Arahy.VMLN7L and Arahy.7XWF6F, with Arahy.VMLN7L negatively regulating seed vigor and Arahy.7XWF6F positively regulating it, suggesting distinct regulatory mechanisms. This study confirms that GWAS based on hyperspectral phenotyping reveals genetic relationships in seed vigor levels, offering novel insights and directions for future peanut breeding, accelerating genetic improvements, and boosting agricultural yields. This approach can be extended to monitor and explore germplasms and other key variables in various crops.

A Dual-Branch Fusion of a Graph Convolutional Network and a Convolutional Neural Network for Hyperspectral Image Classification

Semi-supervised graph convolutional networks (SSGCNs) have been proven to be effective in hyperspectral image classification (HSIC). However, limited training data and spectral uncertainty restrict the classification performance, and the computational demands of a graph convolution network (GCN) present challenges for real-time applications. To overcome these issues, a dual-branch fusion of a GCN and convolutional neural network (DFGCN) is proposed for HSIC tasks. The GCN branch uses an adaptive multi-scale superpixel segmentation method to build fusion adjacency matrices at various scales, which improves the graph convolution efficiency and node representations. Additionally, a spectral feature enhancement module (SFEM) enhances the transmission of crucial channel information between the two graph convolutions. Meanwhile, the CNN branch uses a convolutional network with an attention mechanism to focus on detailed features of local areas. By combining the multi-scale superpixel features from the GCN branch and the local pixel features from the CNN branch, this method leverages complementary features to fully learn rich spatial-spectral information. Our experimental results demonstrate that the proposed method outperforms existing advanced approaches in terms of classification efficiency and accuracy across three benchmark data sets.

Rapid Detection of Cleanliness on Direct Bonded Copper Substrate by Using UV Hyperspectral Imaging

In the manufacturing process of electrical devices, ensuring the cleanliness of technical surfaces, such as direct bonded copper substrates, is crucial. An in-line monitoring system for quality checking must provide sufficiently resolved lateral data in a short time. UV hyperspectral imaging is a promising in-line method for rapid, contactless, and large-scale detection of contamination; thus, UV hyperspectral imaging (225-400 nm) was utilized to characterize the cleanliness of direct bonded copper in a non-destructive way. In total, 11 levels of cleanliness were prepared, and a total of 44 samples were measured to develop multivariate models for characterizing and predicting the cleanliness levels. The setup included a pushbroom imager, a deuterium lamp, and a conveyor belt for laterally resolved measurements of copper surfaces. A principal component analysis (PCA) model effectively differentiated among the sample types based on the first two principal components with approximately 100.0% explained variance. A partial least squares regression (PLS-R) model to determine the optimal sonication time showed reliable performance, with R2cv = 0.928 and RMSECV = 0.849. This model was able to predict the cleanliness of each pixel in a testing sample set, exemplifying a step in the manufacturing process of direct bonded copper substrates. Combined with multivariate data modeling, the in-line UV prototype system demonstrates a significant potential for further advancement towards its application in real-world, large-scale processes.

Enhancing Water-Deficient Potato Plant Identification: Assessing Realistic Performance of Attention-Based Deep Neural Networks and Hyperspectral Imaging for Agricultural Applications

Hyperspectral imaging has emerged as a pivotal technology in agricultural research, offering a powerful means to non-invasively monitor stress factors, such as drought, in crops like potato plants. In this context, the integration of attention-based deep learning models presents a promising avenue for enhancing the efficiency of stress detection, by enabling the identification of meaningful spectral channels. This study assesses the performance of deep learning models on two potato plant cultivars exposed to water-deficient conditions. It explores how various sampling strategies and biases impact the classification metrics by using a dual-sensor hyperspectral imaging systems (VNIR -Visible and Near-Infrared and SWIR-Short-Wave Infrared). Moreover, it focuses on pinpointing crucial wavelengths within the concatenated images indicative of water-deficient conditions. The proposed deep learning model yields encouraging results. In the context of binary classification, it achieved an area under the receiver operating characteristic curve (AUC-ROC-Area Under the Receiver Operating Characteristic Curve) of 0.74 (95% CI: 0.70, 0.78) and 0.64 (95% CI: 0.56, 0.69) for the KIS Krka and KIS Savinja varieties, respectively. Moreover, the corresponding F1 scores were 0.67 (95% CI: 0.64, 0.71) and 0.63 (95% CI: 0.56, 0.68). An evaluation of the performance of the datasets with deliberately introduced biases consistently demonstrated superior results in comparison to their non-biased equivalents. Notably, the ROC-AUC values exhibited significant improvements, registering a maximum increase of 10.8% for KIS Krka and 18.9% for KIS Savinja. The wavelengths of greatest significance were observed in the ranges of 475-580 nm, 660-730 nm, 940-970 nm, 1420-1510 nm, 1875-2040 nm, and 2350-2480 nm. These findings suggest that discerning between the two treatments is attainable, despite the absence of prominently manifested symptoms of drought stress in either cultivar through visual observation. The research outcomes carry significant implications for both precision agriculture and potato breeding. In precision agriculture, precise water monitoring enhances resource allocation, irrigation, yield, and loss prevention. Hyperspectral imaging holds potential to expedite drought-tolerant cultivar selection, thereby streamlining breeding for resilient potatoes adaptable to shifting climates.

A critical systematic review on spectral-based soil nutrient prediction using machine learning

The United Nations (UN) emphasizes the pivotal role of sustainable agriculture in addressing persistent starvation and working towards zero hunger by 2030 through global development. Intensive agricultural practices have adversely impacted soil quality, necessitating soil nutrient analysis for enhancing farm productivity and environmental sustainability. Researchers increasingly turn to Artificial Intelligence (AI) techniques to improve crop yield estimation and optimize soil nutrition management. This study reviews 155 papers published from 2014 to 2024, assessing the use of machine learning (ML) and deep learning (DL) in predicting soil nutrients. It highlights the potential of hyperspectral and multispectral sensors, which enable precise nutrient identification through spectral analysis across multiple bands. The study underscores the importance of feature selection techniques to improve model performance by eliminating redundant spectral bands with weak correlations to targeted nutrients. Additionally, the use of spectral indices, derived from mathematical ratios of spectral bands based on absorption spectra, is examined for its effectiveness in accurately predicting soil nutrient levels. By evaluating various performance measures and datasets related to soil nutrient prediction, this paper offers comprehensive insights into the applicability of AI techniques in optimizing soil nutrition management. The insights gained from this review can inform future research and policy decisions to achieve global development goals and promote environmental sustainability.

Hyperspectral imaging and its applications: A review

Hyperspectral imaging has emerged as an effective powerful tool in plentiful military, environmental, and civil applications over the last three decades. The modern remote sensing approaches are adequate for covering huge earth surfaces with phenomenal temporal, spectral, and spatial resolutions. These features make HSI more effective in various applications of remote sensing depending upon the physical estimation of identical material identification and manifold composite surfaces having accomplished spectral resolutions. Recently, HSI has attained immense significance in the research on safety and quality assessment of food, medical analysis, and agriculture applications. This review focuses on HSI fundamentals and its applications like safety and quality assessment of food, medical analysis, agriculture, water resources, plant stress identification, weed & crop discrimination, and flood management. Various investigators have promising solutions for automatic systems depending upon HSI. Future research may use this review as a baseline and future advancement analysis.

Multichannel Hierarchical Analysis of Time-Resolved Hyperspectral Data for Advanced Colorimetric E-Nose

The colorimetric sensor-based electronic nose has been demonstrated to discriminate specific gaseous molecules for various applications, including health or environmental monitoring. However, conventional colorimetric sensor systems rely on RGB sensors, which cannot capture the complete spectral response of the system. This limitation can degrade the performance of machine learning analysis, leading to inaccurate identification of chemicals with similar functional groups. Here, we propose a novel time-resolved hyperspectral (TRH) data set from colorimetric array sensors consisting of 1D spatial, 1D spectral, and 1D temporal axes, which enables hierarchical analysis of multichannel 2D spectrograms via a convolution neural network (CNN). We assessed the outstanding classification performance of the TRH data set compared to an RGB data set by conducting a relative humidity (RH) concentration classification. The time-dependent spectral response of the colorimetric sensor was measured and trained as a CNN model using TRH and RGB sensor systems at different RH levels. While the TRH model shows a high classification accuracy of 97.5% for the RH concentration, the RGB model yields 72.5% under identical conditions. Furthermore, we demonstrated the detection of various functional volatile gases with the TRH system by using experimental and simulation approaches. The results reveal distinct spectral features from the TRH system, corresponding to changes in the concentration of each substance.

Extraction of water bodies from high-resolution remote sensing imagery based on a deep semantic segmentation network

The precise delineation of urban aquatic features is of paramount importance in scrutinizing water resources, monitoring floods, and devising water management strategies. Addressing the challenge of indistinct boundaries and the erroneous classification of shadowed regions as water in high-resolution remote sensing imagery, we introduce WaterDeep, which is a novel deep learning framework inspired by the DeepLabV3 + architecture and an innovative fusion mechanism for high- and low-level features. This methodology first creates a comprehensive dataset of high-resolution remote sensing images, then progresses through the Xception baseline network for low-level feature extraction, and harnesses densely connected Atrous Spatial Pyramid Pooling (ASPP) modules to assimilate multi-scale data into sophisticated high-level features. Subsequently, the network decoder amalgamates the elemental and intricate features and applies dual-line interpolation to the amalgamated dataset to extract aqueous formations from the remote images. Experimental evidence substantiates that WaterDeep outperforms its existing deep learning counterparts, achieving a stellar overall accuracy of 99.284%, FWIoU of 95.58%, precision of 97.562%, recall of 95.486%, and F1 score of 96.513%. It also excels in the precise demarcation of edges and the discernment of shadows cast by urban infrastructure. The superior efficacy of the proposed method in differentiating water bodies in complex urban environments has significant practical applications in real-world contexts.

Recent advances and challenges in the analysis of natural toxins

Natural toxins (NTs) are poisonous secondary metabolites produced by living organisms developed to ward off predators. Especially low molecular weight NTs (MW<∼1 kDa), such as mycotoxins, phycotoxins, and plant toxins, are considered an important and growing food safety concern. Therefore, accurate risk assessment of food and feed for the presence of NTs is crucial. Currently, the analysis of NTs is predominantly performed with targeted high pressure liquid chromatography tandem mass spectrometry (HPLC-MS/MS) methods. Although these methods are highly sensitive and accurate, they are relatively expensive and time-consuming, while unknown or unexpected NTs will be missed. To overcome this, novel on-site screening methods and non-targeted HPLC high resolution mass spectrometry (HRMS) methods have been developed. On-site screening methods can give non-specialists the possibility for broad "scanning" of potential geographical regions of interest, while also providing sensitive and specific analysis at the point-of-need. Non-targeted chromatography-HRMS methods can detect unexpected as well as unknown NTs and their metabolites in a lab-based approach. The aim of this chapter is to provide an insight in the recent advances, challenges, and perspectives in the field of NTs analysis both from the on-site and the laboratory perspective.

Evaluation of Spectrum-Aided Visual Enhancer (SAVE) in Esophageal Cancer Detection Using YOLO Frameworks

The early detection of esophageal cancer presents a substantial difficulty, which contributes to its status as a primary cause of cancer-related fatalities. This study used You Only Look Once (YOLO) frameworks, specifically YOLOv5 and YOLOv8, to predict and detect early-stage EC by using a dataset sourced from the Division of Gastroenterology and Hepatology, Ditmanson Medical Foundation, Chia-Yi Christian Hospital. The dataset comprised 2741 white-light images (WLI) and 2741 hyperspectral narrowband images (HSI-NBI). They were divided into 60% training, 20% validation, and 20% test sets to facilitate robust detection. The images were produced using a conversion method called the spectrum-aided vision enhancer (SAVE). This algorithm can transform a WLI into an NBI without requiring a spectrometer or spectral head. The main goal was to identify dysplasia and squamous cell carcinoma (SCC). The model's performance was evaluated using five essential metrics: precision, recall, F1-score, mAP, and the confusion matrix. The experimental results demonstrated that the HSI model exhibited improved learning capabilities for SCC characteristics compared with the original RGB images. Within the YOLO framework, YOLOv5 outperformed YOLOv8, indicating that YOLOv5's design possessed superior feature-learning skills. The YOLOv5 model, when used in conjunction with HSI-NBI, demonstrated the best performance. It achieved a precision rate of 85.1% (CI95: 83.2-87.0%, p < 0.01) in diagnosing SCC and an F1-score of 52.5% (CI95: 50.1-54.9%, p < 0.01) in detecting dysplasia. The results of these figures were much better than those of YOLOv8. YOLOv8 achieved a precision rate of 81.7% (CI95: 79.6-83.8%, p < 0.01) and an F1-score of 49.4% (CI95: 47.0-51.8%, p < 0.05). The YOLOv5 model with HSI demonstrated greater performance than other models in multiple scenarios. This difference was statistically significant, suggesting that the YOLOv5 model with HSI significantly improved detection capabilities.

Current and Near-Term Earth-Observing Environmental Satellites, Their Missions, Characteristics, Instruments, and Applications

Among the essential tools to address global environmental information requirements are the Earth-Observing (EO) satellites with free and open data access. This paper reviews those EO satellites from international space programs that already, or will in the next decade or so, provide essential data of importance to the environmental sciences that describe Earth's status. We summarize factors distinguishing those pioneering satellites placed in space over the past half century, and their links to modern ones, and the changing priorities for spaceborne instruments and platforms. We illustrate the broad sweep of instrument technologies useful for observing different aspects of the physio-biological aspects of the Earth's surface, spanning wavelengths from the UV-A at 380 nanometers to microwave and radar out to 1 m. We provide a background on the technical specifications of each mission and its primary instrument(s), the types of data collected, and examples of applications that illustrate these observations. We provide websites for additional mission details of each instrument, the history or context behind their measurements, and additional details about their instrument design, specifications, and measurements.

How to Identify and Quantify Microplastics and Nanoplastics Using Raman Imaging?

While microplastics and nanoplastics are emerging as a big environmental concern, their characterization is still a challenge, particularly for identification and simultaneous quantification analysis where imaging via a hyper spectrum is generally needed. In the past few years, Raman imaging has been greatly advanced, but the analysis protocol is complicated and not yet standardized because imaging analysis is different from traditional analysis. Herein we provide a step-by-step demonstration of how to employ confocal Raman techniques to image microplastics and nanoplastics.

Detection of sweet corn seed viability based on hyperspectral imaging combined with firefly algorithm optimized deep learning

The identification of sweet corn seed vitality is an essential criterion for selecting high-quality varieties. In this research, a combination of hyperspectral imaging technique and diverse deep learning algorithms has been utilized to identify different vitality grades of sweet corn seeds. First, the hyperspectral data of 496 seeds, including four viability-grade seeds, are extracted and preprocessed. Then, support vector machine (SVM) and extreme learning machine (ELM) are used to construct the classification models. Finally, the one-dimensional convolutional neural networks (1DCNN), one-dimensional long short-term memory (1DLSTM), the CNN combined with the LSTM (CNN-LSTM), and the proposed firefly algorithm (FA) optimized CNN-LSTM (FA-CNN-LSTM) are utilized to distinguish spectral images of sweet corn seeds viability grade. The findings from the experimental analysis indicate that the deep learning models exhibit a significant advantage over traditional machine learning approaches in the discrimination of seed vitality levels, boasting a classification accuracy exceeding 94.26% in test datasets and achieving an accuracy improvement of at least 3% compared to the best-performing machine learning model. Moreover, the performance of the FA-CNN-LSTM model proposed in this study demonstrated a slight superiority over the other three models. Besides, the FA-CNN-LSTM achieved a classification accuracy of 97.23%, representing a significant improvement of 2.97% compared to the lowest-performing CNN and a 1.49% enhancement over the CNN-LSTM. In summary, this study reveals the potential of integrating deep learning with hyperspectral imaging as a promising alternative for discriminating sweet corn seed vitality grade, showcasing its value in agricultural research and cultivar breeding.

AUD-Net: A Unified Deep Detector for Multiple Hyperspectral Image Anomaly Detection via Relation and Few-Shot Learning

This article addresses the problem of the building an out-of-the-box deep detector, motivated by the need to perform anomaly detection across multiple hyperspectral images (HSIs) without repeated training. To solve this challenging task, we propose a unified detector [anomaly detection network (AUD-Net)] inspired by few-shot learning. The crucial issues solved by AUD-Net include: how to improve the generalization of the model on various HSIs that contain different categories of land cover; and how to unify the different spectral sizes between HSIs. To achieve this, we first build a series of subtasks to classify the relations between the center and its surroundings in the dual window. Through relation learning, AUD-Net can be more easily generalized to unseen HSIs, as the relations of the pixel pairs are shared among different HSIs. Secondly, to handle different HSIs with various spectral sizes, we propose a pooling layer based on the vector of local aggregated descriptors, which maps the variable-sized features to the same space and acquires the fixed-sized relation embeddings. To determine whether the center of the dual window is an anomaly, we build a memory model by the transformer, which integrates the contextual relation embeddings in the dual window and estimates the relation embeddings of the center. By computing the feature difference between the estimated relation embeddings of the centers and the corresponding real ones, the centers with large differences will be detected as anomalies, as they are more difficult to be estimated by the corresponding surroundings. Extensive experiments on both the simulation dataset and 13 real HSIs demonstrate that this proposed AUD-Net has strong generalization for various HSIs and achieves significant advantages over the specific-trained detectors for each HSI.

Non-destructive assessment of cannabis quality during drying process using hyperspectral imaging and machine learning

Cannabis sativa L. is an industrially valuable plant known for its cannabinoids, such as cannabidiol (CBD) and Δ9-tetrahydrocannabinol (THC), renowned for its therapeutic and psychoactive properties. Despite its significance, the cannabis industry has encountered difficulties in guaranteeing consistent product quality throughout the drying process. Hyperspectral imaging (HSI), combined with advanced machine learning technology, has been used to predict phytochemicals that presents a promising solution for maintaining cannabis quality control. We examined the dynamic changes in cannabinoid compositions under diverse drying conditions and developed a non-destructive method to appraise the quality of cannabis flowers using HSI and machine learning. Even when the relative weight and water content remained constant throughout the drying process, drying conditions significantly influenced the levels of CBD, THC, and their precursors. These results emphasize the importance of determining the exact drying endpoint. To develop HSI-based models for predicting cannabis quality indicators, including dryness, precursor conversion of CBD and THC, and CBD : THC ratio, we employed various spectral preprocessing methods and machine learning algorithms, including logistic regression (LR), support vector machine (SVM), k-nearest neighbor (KNN), random forest (RF), and Gaussian naïve Bayes (GNB). The LR model demonstrated the highest accuracy at 94.7-99.7% when used in conjunction with spectral pre-processing techniques such as multiplicative scatter correction (MSC) or Savitzky-Golay filter. We propose that the HSI-based model holds the potential to serve as a valuable tool for monitoring cannabinoid composition and determining optimal drying endpoint. This tool offers the means to achieve uniform cannabis quality and optimize the drying process in the industry.

Application of Hyperspectral Technology with Machine Learning for Brix Detection of Pastry Pears

Sugar content is an essential indicator for evaluating crisp pear quality and categorization, being used for fruit quality identification and market sales prediction. In this study, we paired a support vector machine (SVM) algorithm with genetic algorithm optimization to reliably estimate the sugar content in crisp pears. We evaluated the spectral data and actual sugar content in crisp pears, then applied three preprocessing methods to the spectral data: standard normal variable transformation (SNV), multivariate scattering correction (MSC), and convolution smoothing (SG). Support vector regression (SVR) models were built using processing approaches. According to the findings, the SVM model preprocessed with convolution smoothing (SG) was the most accurate, with a correlation coefficient 0.0742 higher than that of the raw spectral data. Based on this finding, we used competitive adaptive reweighting (CARS) and the continuous projection algorithm (SPA) to select key representative wavelengths from the spectral data. Finally, we used the retrieved characteristic wavelength data to create a support vector machine model (GASVR) that was genetically tuned. The correlation coefficient of the SG-GASVR model in the prediction set was higher by 0.0321 and the root mean square prediction error (RMSEP) was lower by 0.0267 compared with those of the SG-SVR model. The SG-CARS-GASVR model had the highest correlation coefficient, at 0.8992. In conclusion, the developed SG-CARS-GASVR model provides a reliable method for detecting the sugar content in crisp pear using hyperspectral technology, thereby increasing the accuracy and efficiency of the quality assessment of crisp pear.

Quantitative measurement of internal quality of carrots using hyperspectral imaging and multivariate analysis

The study aimed to measure the carotenoid (Car) and pH contents of carrots using hyperspectral imaging. A total of 300 images were collected using a hyperspectral imaging system, covering 472 wavebands from 400 to 1000 nm. Regions of interest (ROIs) were defined to extract average spectra from the hyperspectral images (HIS). We developed two models: least squares support vector machine (LS-SVM) and partial least squares regression (PLSR) to establish a quantitative analysis between the pigment amounts and spectra. The spectra and pigment contents were predicted and correlated using these models. The selection of EWs for modeling was done using the Successive Projections Algorithm (SPA), regression coefficients (RC) from PLSR models, and LS-SVM. The results demonstrated that hyperspectral imaging could effectively evaluate the internal attributes of carrot cortex and xylem. Moreover, these models accurately predicted the Car and pH contents of the carrot parts. This study provides a valuable approach for variable selection and modeling in hyperspectral imaging studies of carrots.

Unveiling the potential of diffusion model-based framework with transformer for hyperspectral image classification

Hyperspectral imaging has gained popularity for analysing remotely sensed images in various fields such as agriculture and medical. However, existing models face challenges in dealing with the complex relationships and characteristics of spectral-spatial data due to the multi-band nature and data redundancy of hyperspectral data. To address this limitation, we propose a novel approach called DiffSpectralNet, which combines diffusion and transformer techniques. The diffusion method is able extract diverse and meaningful spectral-spatial features, leading to improvement in HSI classification. Our approach involves training an unsupervised learning framework based on the diffusion model to extract high-level and low-level spectral-spatial features, followed by the extraction of intermediate hierarchical features from different timestamps for classification using a pre-trained denoising U-Net. Finally, we employ a supervised transformer-based classifier to perform the HSI classification. We conduct comprehensive experiments on three publicly available datasets to validate our approach. The results demonstrate that our framework significantly outperforms existing approaches, achieving state-of-the-art performance. The stability and reliability of our approach are demonstrated across various classes in all datasets.

Lossless and Near-Lossless Compression Algorithms for Remotely Sensed Hyperspectral Images

Rapid and continuous advancements in remote sensing technology have resulted in finer resolutions and higher acquisition rates of hyperspectral images (HSIs). These developments have triggered a need for new processing techniques brought about by the confined power and constrained hardware resources aboard satellites. This article proposes two novel lossless and near-lossless compression methods, employing our recent seed generation and quadrature-based square rooting algorithms, respectively. The main advantage of the former method lies in its acceptable complexity utilizing simple arithmetic operations, making it suitable for real-time onboard compression. In addition, this near-lossless compressor could be incorporated for hard-to-compress images offering a stabilized reduction at nearly 40% with a maximum relative error of 0.33 and a maximum absolute error of 30. Our results also show that a lossless compression performance, in terms of compression ratio, of up to 2.6 is achieved when testing with hyperspectral images from the Corpus dataset. Further, an improvement in the compression rate over the state-of-the-art k2-raster technique is realized for most of these HSIs by all four variations of our proposed lossless compression method. In particular, a data reduction enhancement of up to 29.89% is realized when comparing their respective geometric mean values.

HyperColorization: propagating spatially sparse noisy spectral clues for reconstructing hyperspectral images

Hyperspectral cameras face challenging spatial-spectral resolution trade-offs and are more affected by shot noise than RGB photos taken over the same total exposure time. Here, we present a colorization algorithm to reconstruct hyperspectral images from a grayscale guide image and spatially sparse spectral clues. We demonstrate that our algorithm generalizes to varying spectral dimensions for hyperspectral images, and show that colorizing in a low-rank space reduces compute time and the impact of shot noise. To enhance robustness, we incorporate guided sampling, edge-aware filtering, and dimensionality estimation techniques. Our method surpasses previous algorithms in various performance metrics, including SSIM, PSNR, GFC, and EMD, which we analyze as metrics for characterizing hyperspectral image quality. Collectively, these findings provide a promising avenue for overcoming the time-space-wavelength resolution trade-off by reconstructing a dense hyperspectral image from samples obtained by whisk or push broom scanners, as well as hybrid spatial-spectral computational imaging systems.

A method to assess industrial paraffin contamination levels in rice and its transferability analysis based on transfer component analysis

Accurate assessment of industrial paraffin contamination levels (IPCLs) in rice is critical for food safety. However, time-consuming and labor-intensive experiments to produce labels for targeted adulterated rice have hindered the development of IPCL estimation methods. In this paper, a transfer learning method (TCA-LSSVR) has been developed. The algorithm integrates transfer component analysis (TCA) with domain adaptive capabilities to produce accurate estimates. Rice from 7 different regions and 3 industrial paraffins were used to generate 4,680 samples from 9 datasets for benchmarking. The test results showed that the established algorithm achieved good estimation performance in various modelling strategies, and only 20 % of off-site samples were needed to supplement the source dataset, the average determination coefficient R2 reached 0.7045, the average RMSE reached 0.140 %, and the average RPD reached 2.023. This work highlights the prospect of rapidly developing a new generation of adulteration detection algorithms using only previous trial data.

Estimation of daylight spectral power distribution from uncalibrated hyperspectral radiance images

This paper introduces a novel framework for estimating the spectral power distribution of daylight illuminants in uncalibrated hyperspectral images, particularly beneficial for drone-based applications in agriculture and forestry. The proposed method uniquely combines image-dependent plausible spectra with a database of physically possible spectra, utilizing an image-independent principal component space (PCS) for estimations. This approach effectively narrows the search space in the spectral domain and employs a random walk methodology to generate spectral candidates, which are then intersected with a pre-trained PCS to predict the illuminant. We demonstrate superior performance compared to existing statistics-based methods across various metrics, validating the framework's efficacy in accurately estimating illuminants and recovering reflectance values from radiance data. The method is validated within the spectral range of 382-1002 nm and shows potential for extension to broader spectral ranges.

Non-destructive detection of egg white and yolk morphology transformation and salt content of salted duck eggs in salting by hyperspectral imaging

Salt content is a crucial indicator of the maturity and internal quality of salted duck eggs (SDEs) during the pickling process. However, there is currently no valid and rapid method available for accurately detecting salt content. In the present study, we utilized hyperspectral imaging to no-destructively determine the salt content in egg yolks, egg whites, and whole eggs during the curing period. Firstly, principal component analysis was applied to explain the characteristics of egg yolk and white morphology transformation of SDEs with different maturities during curing. Secondly, sensitive spectral factors representative of changes in the salt content of SDEs were extracted by three spectral transformations (Savitzky-Golay SG, continuum removal CR, and first-order derivation FD) and three approaches of selecting characteristic wavelengths (successive projection algorithm SPA, uninformative variables elimination UVE and competitive adaptive reweighting sampling algorithm CARS). The results of the PLSR model suggested that the optimal models for predicting salt content in egg yolks, whites, and whole eggs were SG-UVE-PLSR (predicted coefficient of determination Rp2=0.912, predicted standard deviation SEp=0.151, residual prediction deviation RPD = 3.371), CR-CARS-PLSR (Rp2=0.873, SEp=0.862, RPD = 2.806), and CR-UVE-PLSR (Rp2=0.877, SEp=0.680, RPD = 2.851), respectively. Eventually, the optimal prediction model for the salt content of the whole egg was employed to a pixel spectral matrix to calculate the salt content values of pixel points on the hyperspectral image of SDEs. Additionally, pseudo-color techniques were employed to visualize the spatial distribution of predicted salt content. This work will provide a theoretical foundation for rapidly detecting maturity and enabling high-throughput quality sorting of SDEs.

Planar 16-band metasurface-enhanced spectral filter for integrated image sensing

We study theoretically and demonstrate experimentally a 16-band narrow band wavelength selective filter in the near-infrared range. The combination of a pair of distributed Bragg reflectors with a sub-wavelength grating metasurface embedded in the intra-cavity provides a narrow response which can be tuned by adjusting the geometry of the sub-wavelength grating metasurface. The key advantage of this approach is its ease of fabrication, where the spectral response is tuned by merely changing the grating period, resulting in a perfectly planar geometry that can be easily integrated with a broad variety of photodetectors, thus enabling attractive applications such as bio-imaging, time-of-flight sensors and LiDAR. The experimental results are supported by numerical simulations and effective medium theory that unveil the mechanisms that lead to the optical response of the device. It is also shown how the polarization dependence of the structure can be used to determine very accurately the polarization of incoming light.

From celebration to contamination: Analysing microplastics released by burst balloons

Air balloons are a ubiquitous presence in our daily lives, and their rupture may release a substantial quantity of debris, as investigated herein. We employ Raman imaging to capture the fragments resulting from balloon explosions, enabling the identification and direct visualisation of minute microplastic particles / fragments with an improved signal-to-noise ratio for precise quantification. To circumvent the generation of misleading confocal Raman images, we recommend employing terrain mapping to scan the three-dimensional surface of the sample. It is important to acknowledge that the analysis of microplastics at the micro-scale inherently poses limitations in terms of throughput, as it necessitates a trade-off between low and high magnifications. We conduct explosive experiments on ten-to-hundred balloons, collecting debris from various angles and positions. Our investigation involves the random testing of multiple samples / sample positions at the micro-scale, with subsequent extrapolation to estimate the total amount of microplastics. The amalgamation of these results through statistical analysis indicates that each balloon explosion can potentially release tens-to-thousands of microplastics, highlighting a concern that has hitherto received insufficient attention. The characterisation approach, particularly the random Raman scanning method in combination with SEM and the statistical analysis on accumulated samples employed in this report, has the potential to serve as a useful tool in future research on microplastics and even nanoplastics.

Research Progress on Quality Detection of Livestock and Poultry Meat Based on Machine Vision, Hyperspectral and Multi-Source Information Fusion Technologies

Presently, the traditional methods employed for detecting livestock and poultry meat predominantly involve sensory evaluation conducted by humans, chemical index detection, and microbial detection. While these methods demonstrate commendable accuracy in detection, their application becomes more challenging when applied to large-scale production by enterprises. Compared with traditional detection methods, machine vision and hyperspectral technology can realize real-time online detection of large throughput because of their advantages of high efficiency, accuracy, and non-contact measurement, so they have been widely concerned by researchers. Based on this, in order to further enhance the accuracy of online quality detection for livestock and poultry meat, this article presents a comprehensive overview of methods based on machine vision, hyperspectral, and multi-sensor information fusion technologies. This review encompasses an examination of the current research status and the latest advancements in these methodologies while also deliberating on potential future development trends. The ultimate objective is to provide pertinent information and serve as a valuable research resource for the non-destructive online quality detection of livestock and poultry meat.

Optimal-Band Analysis for Chlorophyll Quantification in Rice Leaves Using a Custom Hyperspectral Imaging System

Hyperspectral imaging (HSI) is a promising tool in chlorophyll quantification, providing a non-invasive method to collect important information for effective crop management. HSI contributes to food security solutions by optimising crop yields. In this study, we presented a custom HSI system specifically designed to provide a quantitative analysis of leaf chlorophyll content (LCC). To ensure precise estimation, significant wavelengths were identified using optimal-band analysis. Our research was centred on two sets of 120 leaf samples sourced from Thailand's unique Chaew Khing rice variant. The samples were subjected to (i) an analytical LCC assessment and (ii) HSI imaging for spectral reflectance data capture. A linear regression comparison of these datasets revealed that the green (575 ± 2 nm) and near-infrared (788 ± 2 nm) bands were the most outstanding performers. Notably, the green normalised difference vegetation index (GNDVI) was the most reliable during cross-validation (R2=0.78 and RMSE = 2.4 µg∙cm-2), outperforming other examined vegetable indices (VIs), such as the simple ratio (RED/GREEN) and the chlorophyll index. The potential development of a streamlined sensor dependent only on these two wavelengths is a significant outcome of identifying these two optimal bands. This innovation can be seamlessly integrated into farming landscapes or attached to UAVs, allowing real-time monitoring and rapid, targeted N management interventions.

Unveiling microplastics from zippers: Characterisation and visualisation through Raman imaging analysis

Microplastics have emerged as a global concern due to the increased plastic contamination found in a variety of sources. Herein we unveil microplastics released from plastic zippers that can generally be found in our clothes and textiles. We first employ a scanning electron microscope (SEM) to visualise the scratches developed on the zipper teeth and the derived particles. We then use Raman imaging to identify and simultaneously visualise the plastics from the chemical or molecular spectrum window. Based on hundreds to thousands of spectra, rather than a single spectrum or even a single peak that works as just a pixel in the image, imaging analysis can significantly increase the signal-to-noise ratio. Furthermore, the non-uniform distribution of components or multi-components can also be effectively imaged to avoid the possible bias from the single-spectrum analysis. The challenge to convert the hundreds to thousands of spectra of a hyperspectral matrix to an image is also discussed, and chemometrics is adopted and recommended to further improve the signal-to-noise ratio. The co-ingredient of titanium oxide in the zipper teeth/sewing lines is also effectively identified by Raman imaging. Based on the effective characterisation, we estimate that up to ~410 microplastics could be potentially released during each time of on-off zipping, although the variation can be expected and depends on several other factors. This study reminds us to be aware of the potential contamination derived from similar types of microplastic sources in our daily lives.

A Novel Correction Methodology to Improve the Performance of a Low-Cost Hyperspectral Portable Snapshot Camera

The development of spectral sensors (SSs) capable of retrieving spectral information have opened new opportunities to improve several environmental and agricultural practices, e.g., crop breeding, plant phenotyping, land use monitoring, and crop classification. The SSs are classified as multispectral and hyperspectral (HS) based on the number of the spectral bands resolved and sampled during data acquisition. Large-scale applications of the HS remain limited due to the cost of this type of technology and the technical difficulties in hyperspectral data processing. Low-cost portable hyperspectral cameras (PHCs) have been progressively developed; however, critical aspects associated with data acquisition and processing, such as the presence of spectral discontinuities, signal jumps, and a high level of background noise, were reported. The aim of this work was to analyze and improve the hyperspectral output of a PHC Senop HSC-2 device by developing a general use methodology. Several signal gaps were identified as falls and jumps across the spectral signatures near 513, 650, and 930 nm, while the dark current signal magnitude and variability associated with instrumental noise showed an increasing trend over time. A data correction pipeline was successfully developed and tested, leading to 99% and 74% reductions in radiance signal jumps identified at 650 and 830 nm, respectively, while the impact of noise on the acquired signal was assessed to be in the range of 10% to 15%. The developed methodology can be effectively applied to other low-cost hyperspectral cameras.

Online small-object anti-fringe sorting of tobacco stem impurities based on hyperspectral superpixels

The use of tobacco stems as raw material for cigarettes reduces cost and improves the flammability of cigarettes. However, various impurities, such as plastic, reduce the purity of tobacco stems, degrade the quality of cigarettes, and endanger the health of smokers. Therefore, the correct classification of tobacco stems and impurities is crucial. This study proposes a method based on hyperspectral image superpixels and the use of light gradient boosting machine (LightGBM) classifier to categorize tobacco stems and impurities. First, the hyperspectral image is segmented using superpixels. Second, the gray-level co-occurrence matrix extracts the texture features of superpixels. Subsequently, an improved LightGBM is applied and trained with the spectral and textural features of superpixels as a classification model. Several experiments were implemented to evaluate the performance of the proposed method. The results show that the classification performance based on superpixels is better than that based on single-pixel points. The classification model based on superpixels (10 × 10 px) achieved the highest impurity recognition rate (93.8%). This algorithm has already been applied to industrial production in cigarette factories. It exhibits considerable potential in overcoming the influence of interference fringes to promote the intelligent industrial application of hyperspectral imaging.

A Review of Recent Advances in Computer-Aided Detection Methods Using Hyperspectral Imaging Engineering to Detect Skin Cancer

Skin cancer, a malignant neoplasm originating from skin cell types including keratinocytes, melanocytes, and sweat glands, comprises three primary forms: basal cell carcinoma (BCC), squamous cell carcinoma (SCC), and malignant melanoma (MM). BCC and SCC, while constituting the most prevalent categories of skin cancer, are generally considered less aggressive compared to MM. Notably, MM possesses a greater capacity for invasiveness, enabling infiltration into adjacent tissues and dissemination via both the circulatory and lymphatic systems. Risk factors associated with skin cancer encompass ultraviolet (UV) radiation exposure, fair skin complexion, a history of sunburn incidents, genetic predisposition, immunosuppressive conditions, and exposure to environmental carcinogens. Early detection of skin cancer is of paramount importance to optimize treatment outcomes and preclude the progression of disease, either locally or to distant sites. In pursuit of this objective, numerous computer-aided diagnosis (CAD) systems have been developed. Hyperspectral imaging (HSI), distinguished by its capacity to capture information spanning the electromagnetic spectrum, surpasses conventional RGB imaging, which relies solely on three color channels. Consequently, this study offers a comprehensive exploration of recent CAD investigations pertaining to skin cancer detection and diagnosis utilizing HSI, emphasizing diagnostic performance parameters such as sensitivity and specificity.

Assessment of hyperspectral imaging and CycleGAN-simulated narrowband techniques to detect early esophageal cancer

The clinical signs and symptoms of esophageal cancer (EC) are often not discernible until the intermediate or advanced phases. The detection of EC in advanced stages significantly decreases the survival rate to below 20%. This study conducts a comparative analysis of the efficacy of several imaging techniques, including white light image (WLI), narrowband imaging (NBI), cycle-consistent adversarial network simulated narrowband image (CNBI), and hyperspectral imaging simulated narrowband image (HNBI), in the early detection of esophageal cancer (EC). In conjunction with Kaohsiung Armed Forces General Hospital, a dataset consisting of 1000 EC pictures was used, including 500 images captured using WLI and 500 images captured using NBI. The CycleGAN model was used to generate the CNBI dataset. Additionally, a novel method for HSI imaging was created with the objective of generating HNBI pictures. The evaluation of the efficacy of these four picture types in early detection of EC was conducted using three indicators: CIEDE2000, entropy, and the structural similarity index measure (SSIM). Results of the CIEDE2000, entropy, and SSIM analyses suggest that using CycleGAN to generate CNBI images and HSI model for creating HNBI images is superior in detecting early esophageal cancer compared to the use of conventional WLI and NBI techniques.

Raman imaging to identify microplastics released from toothbrushes: algorithms and particle analysis

Microplastics are small plastic fragments that are of increasing concern due to their potential impacts on the environment and human health. The source of microplastics is not completely clear and might originate in daily lives such as from toothbrushes. When toothbrushes are used to clean teeth, small plastic debris and fragments can be potentially released into mouths directly or environment indirectly. This study aims to examine the release of microplastics from toothbrushes, using Raman imaging to identify and visualise the plastic debris with an increased signal-noise ratio via hyper-spectrum analysis. Using algorithms to convert the hyper-spectrum to an image, the plastic can be distinguished from the co-formulated titanium oxide particles that are not uniformly distributed along the plastics. The non-uniform distribution can lead to the bias results if a single spectrum analysis is conducted at one position rather than imaging analysis to scan an area. The potential false image originating from the off-focal position for the confocal Raman is overcome using the terrain map to guide the Raman imaging. The imaging analysis balancing between the low magnification to capture the overview and the high magnification to test the details is also discussed. While the release amount of microplastics from the toothbrush is estimated at thousands daily with the expected variation, the results of this study have confirmed the release of microplastics in daily lives. The imaging analysis approach along with algorithm can help to identify the chemical elements of microplastics from the complex background, which can benefit the further research on microplastics towards risk assessment and remediation.

Heterogeneous Regularization-Based Tensor Subspace Clustering for Hyperspectral Band Selection

Band selection (BS) reduces effectively the spectral dimension of a hyperspectral image (HSI) by selecting relatively few representative bands, which allows efficient processing in subsequent tasks. Existing unsupervised BS methods based on subspace clustering are built on matrix-based models, where each band is reshaped as a vector. They encode the correlation of data only in the spectral mode (dimension) and neglect strong correlations between different modes, i.e., spatial modes and spectral mode. Another issue is that the subspace representation of bands is performed in the raw data space, where the dimension is often excessively high, resulting in a less efficient and less robust performance. To address these issues, in this article, we propose a tensor-based subspace clustering model for hyperspectral BS. Our model is developed on the well-known Tucker decomposition. The three factor matrices and a core tensor in our model encode jointly the multimode correlations of HSI, avoiding effectively to destroy the tensor structure and information loss. In addition, we propose well-motivated heterogeneous regularizations (HRs) on the factor matrices by taking into account the important local and global properties of HSI along three dimensions, which facilitates the learning of the intrinsic cluster structure of bands in the low-dimensional subspaces. Instead of learning the correlations of bands in the original domain, a common way for the matrix-based models, our model learns naturally the band correlations in a low-dimensional latent feature space, which is derived by the projections of two factor matrices associated with spatial dimensions, leading to a computationally efficient model. More importantly, the latent feature space is learned in a unified framework. We also develop an efficient algorithm to solve the resulting model. Experimental results on benchmark datasets demonstrate that our model yields improved performance compared to the state-of-the-art.

Lightweight 3D Dense Autoencoder Network for Hyperspectral Remote Sensing Image Classification

The lack of labeled training samples restricts the improvement of Hyperspectral Remote Sensing Image (HRSI) classification accuracy based on deep learning methods. In order to improve the HRSI classification accuracy when there are few training samples, a Lightweight 3D Dense Autoencoder Network (L3DDAN) is proposed. Structurally, the L3DDAN is designed as a stacked autoencoder which consists of an encoder and a decoder. The encoder is a hybrid combination of 3D convolutional operations and 3D dense block for extracting deep features from raw data. The decoder composed of 3D deconvolution operations is designed to reconstruct data. The L3DDAN is trained by unsupervised learning without labeled samples and supervised learning with a small number of labeled samples, successively. The network composed of the fine-tuned encoder and trained classifier is used for classification tasks. The extensive comparative experiments on three benchmark HRSI datasets demonstrate that the proposed framework with fewer trainable parameters can maintain superior performance to the other eight state-of-the-art algorithms when there are only a few training samples. The proposed L3DDAN can be applied to HRSI classification tasks, such as vegetation classification. Future work mainly focuses on training time reduction and applications on more real-world datasets.

Review of technology advances to assess rice quality traits and consumer perception

The increase in rice consumption and demand for high-quality rice is impacted by the growth of socioeconomic status in developing countries and consumer awareness of the health benefits of rice consumption. The latter aspects drive the need for rapid, low-cost, and reliable quality assessment methods to produce high-quality rice according to consumer preference. This is important to ensure the sustainability of the rice value chain and, therefore, accelerate the rice industry toward digital agriculture. This review article focuses on the measurements of the physicochemical and sensory quality of rice, including new and emerging technology advances, particularly in the development of low-cost, non-destructive, and rapid digital sensing techniques to assess rice quality traits and consumer perceptions. In addition, the prospects for potential applications of emerging technologies (i.e., sensors, computer vision, machine learning, and artificial intelligence) to assess rice quality and consumer preferences are discussed. The integration of these technologies shows promising potential in the forthcoming to be adopted by the rice industry to assess rice quality traits and consumer preferences at a lower cost, shorter time, and more objectively compared to the traditional approaches.

Enhancing Feature Detection and Matching in Low-Pixel-Resolution Hyperspectral Images Using 3D Convolution-Based Siamese Networks

Today, hyperspectral imaging plays an integral part in the remote sensing and precision agriculture field. Identifying the matching key points between hyperspectral images is an important step in tasks such as image registration, localization, object recognition, and object tracking. Low-pixel resolution hyperspectral imaging is a recent introduction to the field, bringing benefits such as lower cost and form factor compared to traditional systems. However, the use of limited pixel resolution challenges even state-of-the-art feature detection and matching methods, leading to difficulties in generating robust feature matches for images with repeated textures, low textures, low sharpness, and low contrast. Moreover, the use of narrower optics in these cameras adds to the challenges during the feature-matching stage, particularly for images captured during low-altitude flight missions. In order to enhance the robustness of feature detection and matching in low pixel resolution images, in this study we propose a novel approach utilizing 3D Convolution-based Siamese networks. Compared to state-of-the-art methods, this approach takes advantage of all the spectral information available in hyperspectral imaging in order to filter out incorrect matches and produce a robust set of matches. The proposed method initially generates feature matches through a combination of Phase Stretch Transformation-based edge detection and SIFT features. Subsequently, a 3D Convolution-based Siamese network is utilized to filter out inaccurate matches, producing a highly accurate set of feature matches. Evaluation of the proposed method demonstrates its superiority over state-of-the-art approaches in cases where they fail to produce feature matches. Additionally, it competes effectively with the other evaluated methods when generating feature matches in low-pixel resolution hyperspectral images. This research contributes to the advancement of low pixel resolution hyperspectral imaging techniques, and we believe it can specifically aid in mosaic generation of low pixel resolution hyperspectral images.

Performance of finite-size metal-dielectric nanoslits metasurface optical filters

In this work, we analyze the performance of finite-size metal-dielectric nanoslits guided mode resonance metasurface optical filters by using finite-difference time-domain simulations and spatial Fourier transform analysis. It is shown that in the direction of the nanoslits period, the critical size required to maintain the performance of the corresponding infinite size filter is the product of the nanoslits period and the quality factor of the infinite size filter. Size reduction in this dimension below the critical dimension reduces the peak transmittance and broadens the spectral linewidth of the filter. In the dimension orthogonal to the nanoslits period direction, the critical dimension size required is not related to the quality factor of the corresponding infinite size filter. Our analysis shows that the critical size is 12 times the filter peak wavelength in the orthogonal dimension for maintaining the filter performance. The 12 times filter wavelength requirement corresponds to the second zero of the Fourier transform of the aperture function.

Blood cancer diagnosis using hyperspectral imaging combined with the forward searching method and machine learning

Hyperspectral imaging (HSI), a widely used biosensing technique, has been applied to tumor detection. Rapid, accurate, and low-cost detection of blood cancer using hyperspectral technology remains a challenge. We developed a new method to discriminate blood cancer using hyperspectral imaging (HSI) and the forward searching method (FSM). Four commonly used classification models are applied for four types of blood cancer spectra recognition. The support vector machine (SVM) model with the highest recognition accuracy (94.5%) combined with HSI achieves high-precision tumor identification. For higher recognition accuracy and lower hardware barriers, based on the selection probabilities of spectral lines calculated by a multi-objective atomic orbital search method, the FSM is proposed for HSI feature selection. With the proposed method, the wavelength band range of the spectrum is reduced by at least 50%. Compared with the traditional dimensionality reduction methods, the FSM can obtain a higher accuracy rate with lower hardware requirements. These results show that our proposed method can achieve non-invasive rapid screening of blood cancers with lower hardware requirements. Therefore, HSI assisted with FSM and SVM hybrid models can be a powerful and promising tool for blood cancer detection.

Analysis of Hyperspectral Data to Develop an Approach for Document Images

Hyperspectral data analysis is being utilized as an effective and compelling tool for image processing, providing unprecedented levels of information and insights for various applications. In this manuscript, we have compiled and presented a comprehensive overview of recent advances in hyperspectral data analysis that can provide assistance for the development of customized techniques for hyperspectral document images. We review the fundamental concepts of hyperspectral imaging, discuss various techniques for data acquisition, and examine state-of-the-art approaches to the preprocessing, feature extraction, and classification of hyperspectral data by taking into consideration the complexities of document images. We also explore the possibility of utilizing hyperspectral imaging for addressing critical challenges in document analysis, including document forgery, ink age estimation, and text extraction from degraded or damaged documents. Finally, we discuss the current limitations of hyperspectral imaging and identify future research directions in this rapidly evolving field. Our review provides a valuable resource for researchers and practitioners working on document image processing and highlights the potential of hyperspectral imaging for addressing complex challenges in this domain.

The role of remote sensing in tropical grassland nutrient estimation: a review

The carbon (C) and nitrogen (N) ratio is a key indicator of nutrient utilization and limitations in rangelands. To understand the distribution of herbivores and grazing patterns, information on grass quality and quantity is important. In heterogeneous environments, remote sensing offers a timely, economical, and effective method for assessing foliar biochemical ratios at varying spatial and temporal scales. Hence, this study provides a synopsis of the advancement in remote sensing technology, limitations, and emerging opportunities in mapping the C:N ratio in rangelands. Specifically, the paper focuses on multispectral and hyperspectral sensors and investigates their properties, absorption features, empirical and physical methods, and algorithms in predicting the C:N ratio in grasslands. Literature shows that the determination of the C:N ratio in grasslands is not in line with developments in remote sensing technologies. Thus, the use of advanced and freely available sensors with improved spectral and spatial properties such as Sentinel 2 and Landsat 8/9 with sophisticated algorithms may provide new opportunities to estimate C:N ratio in grasslands at regional scales, especially in developing countries. Spectral bands in the near-infrared, shortwave infrared, red, and red edge were identified to predict the C:N ratio in plants. New indices developed from recent multispectral satellite imagery, for example, Sentinel 2 aided by cutting-edge algorithms, can improve the estimation of foliar biochemical ratios. Therefore, this study recommends that future research should adopt new satellite technologies with recent development in machine learning algorithms for improved mapping of the C:N ratio in grasslands.