Hyperspectral Imaging Method Research Articles

Semi-Supervised Tree Species Classification for Multi-Source Remote Sensing Images Based on a Graph Convolutional Neural Network

As a current research hotspot, graph convolution networks (GCNs) have provided new opportunities for tree species classification in multi-source remote sensing images. To solve the challenge of limited label information, a new tree species classification model was proposed by using the semi-supervised graph convolution fusion method for hyperspectral images (HSIs) and multispectral images (MSIs). In the model, the graph-based attribute features and pixel-based features are fused to deepen the correlation of multi-source images to improve accuracy. Firstly, the model employs the canonical correlation analysis (CCA) method to maximize the correlation of multi-source images, which explores the relationship between information from various sources further and offers more valuable insights. Secondly, convolution calculations were made to extract features and then map graph node fusion, which not only reduces redundancy features but also enhances compelling features. Finally, the relationship between representative descriptors is captured through the use of hyperedge convolution in the training process, and the dominant features on the graph are fully mined. The tree species are classified through two fusion feature operations, leading to improved classification performance compared to state-of-the-art methods. The fusion strategy can produce a complete classification map of the study areas.

Investigation into maize seed disease identification based on deep learning and multi-source spectral information fusion techniques

Detection of diseases in maize seeds is crucial for their quality evaluation and disease control. This study uses hyperspectral imaging (HSI) and deep learning methods for analysis and identification. Successive projections algorithm (SPA) and principal component analysis (PCA) were applied to extract feature variables, and data-level fusion, feature-level fusion, and decision-level fusion were employed to process different types of feature data. Classification models with different fusion strategies were built using partial least squares discriminant analysis (PLS-DA), random forest (RF), support vector machine (SVM), and convolutional neural network (CNN-RB). The results show that the modeling performance based on spectral features outperforms that based on color and texture features. Among them, the accuracy of CNN-RB based on feature variable modeling was 94.44 %, which was better than RF (93.89 %) and SVM (92.78 %), and only second to PLS-DA (97.78 %). Different fusion strategies had different performances, among which the decision-level fusion had the best effect, with an accuracy of 98.12 %. The study shows that the proposed CNN-RB model can effectively enhance the feature extraction ability of the network, and the multi-source information fusion technique can improve the recognition performance of the model. The method has great potential for application in seed disease detection.

Optical Identification of Diabetic Retinopathy Using Hyperspectral Imaging.

The severity of diabetic retinopathy (DR) is directly correlated to changes in both the oxygen utilization rate of retinal tissue as well as the blood oxygen saturation of both arteries and veins. Therefore, the current stage of DR in a patient can be identified by analyzing the oxygen content in blood vessels through fundus images. This enables medical professionals to make accurate and prompt judgments regarding the patient's condition. However, in order to use this method to implement supplementary medical treatment, blood vessels under fundus images need to be determined first, and arteries and veins then need to be differentiated from one another. Therefore, the entire study was split into three sections. After first removing the background from the fundus images using image processing, the blood vessels in the images were then separated from the background. Second, the method of hyperspectral imaging (HSI) was utilized in order to construct the spectral data. The HSI algorithm was utilized in order to perform analysis and simulations on the overall reflection spectrum of the retinal image. Thirdly, principal component analysis (PCA) was performed in order to both simplify the data and acquire the major principal components score plot for retinopathy in arteries and veins at all stages. In the final step, arteries and veins in the original fundus images were separated using the principal components score plots for each stage. As retinopathy progresses, the difference in reflectance between the arteries and veins gradually decreases. This results in a more difficult differentiation of PCA results in later stages, along with decreased precision and sensitivity. As a consequence of this, the precision and sensitivity of the HSI method in DR patients who are in the normal stage and those who are in the proliferative DR (PDR) stage are the highest and lowest, respectively. On the other hand, the indicator values are comparable between the background DR (BDR) and pre-proliferative DR (PPDR) stages due to the fact that both stages exhibit comparable clinical-pathological severity characteristics. The results indicate that the sensitivity values of arteries are 82.4%, 77.5%, 78.1%, and 72.9% in the normal, BDR, PPDR, and PDR, while for veins, these values are 88.5%, 85.4%, 81.4%, and 75.1% in the normal, BDR, PPDR, and PDR, respectively.

Coating Condition Detection and Assessment on the Steel Girder of a Bridge through Hyperspectral Imaging

The organic coating of bridge steel girders is subjected to physical scratches, corrosion, and aging in natural weathering. The breakdown of the coating may cause serviceability and safety problems if left unnoticed. Conventional coating inspection is time-consuming and lacks information about the coating’s chemical integrity. A hyperspectral imaging method is proposed to detect the condition of steel coatings based on coating-responsive features in reflectance spectra. A field test was conducted on the real-world bridge, which shows obvious signs of degradation. The hyperspectral signature enables an assessment of the coating’s health and defect severity. The results indicated that the coating scratch can be effectively located in the domain of a hyperspectral image and the scratch depth can be determined by mapping a scratch depth indicator (SDI = R532 nm/R641 nm). Rust sources and products in steel girders can be identified by the unique spectral signatures in the VNIR range, and the rust stains (and thus stain areas) scattered on the coating can be pinpointed at pixel level by the chloride rust (CR) indicators >1.11 (CR = R733 nm/R841 nm). The chemical integrity of a topcoat is demonstrated by the short-wave infrared spectroscopy and the topcoat degradation can be evaluated by the decreased absorption at 8000 cm−1 and 5850 cm−1. Hyperspectral imaging enables faster and more reliable coating condition detection by the spectral features and provides an alternative for multi-object coating detection.

The assessment of moderate acne vulgaris face skin using blood perfusion and hyperspectral imaging-A pilot study.

Microcirculation is the flow of blood through the smallest vessels in the circulatory system. Capillaries respond to various pathologies much earlier than arteries and veins, the changes which indicate that the disease is already developing. Careful monitoring of the functioning of the capillary system often allows the detection of in vivo disorders at an early stage. The aim of the study was to evaluate the microcirculation within acne lesions. A noninvasive method of semi-quantitative blood perfusion measurement based on laser speckle contrast analysis (LASCA) and a quantitative assessment of the acne severity using the hyperspectral imaging method supported by image analysis and processing methods were used. The acne lesion perfusion was compared to the healthy skin perfusion of the control group. The reflectance in the range from 400 to 1000 nm was also compared for acne lesions and healthy skin. A dedicated version of the algorithm based on the gray-level co-occurrence matrix was used to compare differences between acne lesions and healthy skin. Blood perfusion in acne lesions is on average 117% higher than in healthy skin. The reflectance of acne lesions was lower in the range of 400-600 nm compared to the healthy skin. A statistically significant decrease in the reflectance of acne lesions was demonstrated for wavelengths: 434, 549, and 588 nm. Increased blood perfusion in acne lesions compared to healthy skin suggests increased chemotaxis of inflammatory cells and wound healing. Decrease points in skin reflectance resemble the absorption peaks for hemoglobin and coproporphyrin III produced by Cutibacterium acnes. Thus, the proposed methods may serve not only for the early detection of acne lesions, before they manifest themselves clinically for the naked eye, but may also be a tool for quantifying the effectiveness of acne treatment.

The Impact of Water Availability on the Discriminative Status of Nitrogen (N) in Sugar Beet and Celery Using Hyperspectral Imaging Methods

A pot experiment was conducted to determine the impact of water availability on the discriminatory status of nitrogen (N) in plants using hyperspectral imaging. Nitrogen deficiency causes a significant decrease in chlorophyll concentration in plant leaves regardless of water availability. Five different classification algorithms were used to discriminate between nitrogen concentrations in plants at different levels of water availability. Several statistical parameters, including kappa and overall classification accuracy for calibration and prediction, were used to determine the efficiency and accuracy of the models. The Random Forest model had the highest overall accuracy of over 81% for sugar beet and over 78% for celery. Additionally, characteristic electromagnetic wavelengths were identified in which reflectance correlated with nitrogen and water content in plants could be recorded. It was also noted that the spectral resolution between the N and High Water (HW)/Low Water (LW) treatments was lower in the short-wave infrared (SWIR) region than in the visible and near-infrared (VNIR) region.

Classification between live and dead foodborne bacteria with hyperspectral microscope imagery and machine learning

Identifying live foodborne bacteria is essential for ensuring food safety and preventing foodborne illnesses. This study investigated the use of hyperspectral microscope imaging and deep learning methods to accurately distinguish between live and dead foodborne bacteria based on their spectral and morphological features. Three deep learning models, Fusion-Net I, II, and III, were developed and evaluated for their ability to classify live and dead bacterial cells of six pathogenic strains, including Escherichia coli (EC), Listeria innocua (LI), Staphylococcus aureus (SA), Salmonella Enteritidis (SE), Salmonella Heidelberg (SH), and Salmonella Typhimurium (ST). The models utilized both morphological and spectral characteristics of the bacterial cells, with inputs of average spectra and 546 nm band images. Fusion-Net I achieved high accuracy in identifying live bacterial cells, with a classification accuracy of 100% for LI, SE, ST strains and over 92.9% for EC, SA, SH. Fusion-Net II and III models were even more robust, achieving 100% accuracy consistently in classifying dead cells in all six strains. Fusion-Net III also showed the ability to identify bacterial strains with 96.9% accuracy, making it a dual-task model with potential applications in identifying live foodborne bacteria prior to foodborne outbreaks. These findings suggest that the use of hyperspectral microscope imaging and deep learning methods could provide a new tool for quickly and accurately identifying bacterial viability, thereby improving the efficiency and reliability of food safety inspection.

Near infrared hyperspectral imaging method to assess Fusarium Head Blight infection on winter wheat ears

The scope of this article is to present a laboratory-based method for the assessment of Fusarium Head Blight (FHB) on ears of winter wheat using Near Infrared Hyperspectral Imaging (NIR-HSI) in the spectral range 900–1700 nm. Inoculated and non-inoculated winter wheat ears were collected from a trial field in Belgium and analysed in the laboratory. Using a dichotomous classification tree framework, Partial Least Squares Discriminant Analysis (PLS-DA) models were developed and applied at pixel level to discriminate the different parts of the ears, namely the awns, the stem and the healthy and infected parts of the ears. This study proposes the use of spectral information at pixel level instead of averaging the spectral signature of the ear. This approach allows using the spatial information of the data in order to provide a quantitative assessment of the severity of the infection on the ear. The results presented in this work confirm that wavelengths related to water and nitrogen are important for the detection of FHB. The validation of the models indicates a good performance for the detection of FHB infection at pixel level with a sensitivity of 96 % and a specificity of 100 %. Furthermore, the validation of the method at ear level shows good performance for the detection of FHB-infected ears with a sensitivity of 99.4 % and a specificity of 91.9 %. The literature reports similar performances, but using equipment with a broader spectral range (400–2500 nm). The present study thus indicates that good discrimination performance can be achieved using a much narrower spectral range. In addition, the obtained results suggest that the method can discriminate three classes of FHB severity, which enables a semi-quantitative assessment of the FHB infection.

Vis-NIR and SWIR hyperspectral imaging method to detect bruises in pomegranate fruit.

Fresh pomegranate fruit is susceptible to bruising, a common type of mechanical damage during harvest and at all stages of postharvest handling. Accurate and early detection of such damages in pomegranate fruit plays an important role in fruit grading. This study investigated the detection of bruises in fresh pomegranate fruit using hyperspectral imaging technique. A total of 90 sample of pomegranate fruit were divided into three groups of 30 samples, each representing purposefully induced pre-scanning bruise by dropping samples from 100 cm and 60 cm height on a metal surface. The control has no pre-scanning bruise (no drop). Two hyperspectral imaging setups were examined: visible and near infrared (400 to 1000 nm) and short wavelength infrared (1000 to 2500 nm). Region of interest (ROI) averaged reflectance spectra was implemented to reduce the image data. For all hypercubes a principal components analysis (PCA) based background removal were done prior to segmenting the region of interest (ROI) using the Evince® multi-variate analysis software 2.4.0. Then the average spectrum of the ROI of each sample was computed and transferred to the MATLAB 2022a (The MathWorks, Inc., Mass., USA) for classification. A two-layer feed-forward artificial neural network (ANN) is used for classification. The accuracy of bruise severity classification ranged from 80 to 96.7%. When samples from both bruise severity (Bruise damage induced from a 100cm and 60 cm drop heights respectively) cases were merged, class recognition accuracy were 88.9% and 74.4% for the SWIR and Vis-NIR, respectively. This study implemented the method of selecting out informative bands and disregarding the redundant ones to decreases the data size and dimension. The study developed a more compact classification model by the data dimensionality reduction method. This study demonstrated the potential of using hyperspectral imaging technology in sensing and classification of bruise severity in pomegranate fruit. This work provides the foundation to build a compact and fast multispectral imaging-based device for practical farm and packhouse applications.

Multi-task convolutional neural network for simultaneous monitoring of lipid and protein oxidative damage in frozen-thawed pork using hyperspectral imaging

Lipid and protein oxidation are the main causes of meat deterioration during freezing. Traditional methods using hyperspectral imaging (HSI) need to train multiple independent models to predict multiple attributes, which is complex and time-consuming. In this study, a multi-task convolutional neural network (CNN) model was developed for visible near-infrared HSI data (400–1002 nm) of 240 pork samples treated with different freeze-thaw cycles (0–9 cycles) to evaluate the feasibility of simultaneously monitoring lipid oxidation (thiobarbituric acid reactive substance content) and protein oxidation (carbonyl content) in pork. The performance of the commonly used partial least squares regression (PLSR) model based on the spectra after pre-processing (Standard normal variate, Savitzky-Golay derivative, and Savitzky-Golay smoothing) and feature selection (Regression coefficients) and single-output CNN model was compared. The results showed that the multi-task CNN model achieved the optimal prediction accuracies for lipid oxidation (R2p = 0.9724, RMSEP = 0.0227, and RPD = 5.2579) and protein oxidation (R2p = 0.9602, RMSEP = 0.0702, and RPD = 4.6668). In final, the changes of lipid and protein oxidation of pork in different freeze-thaw cycles were successfully visualized. In conclusion, the combination of HSI and multi-task CNN method shows the potential of end-to-end prediction of pork oxidative damage. This study provides a new, convenient and automated technique for meat quality detection in the food industry.

Visible and near-infrared hyperspectral imaging as an intelligent tool for parasite detection in sashimi

There are parasites found on sashimi which can cause a series of health problems to those who consume them. Because the parasites are too small to be visible to the naked eye, the labor and time-consuming use of a microscope is required for detection. This study proposes a visible and near-infrared (VIS/NIR) hyperspectral imaging method to quickly and intelligently detect parasites in sashimi. The research results show that the VIS/NIR spectrums for fish meat and parasite images were different at certain wavelengths. The ability of a probabilistic neural network (PNN) combined with multiple detection models was better than that of partial least squares regression (PLSR) combined with a single detection model for the true positive detection of parasites on sashimi. A synthesis between PNN and a combination of detection models, including Savitzky-Golay, standard normal variate, and first derivative pre-processing, is able to optimally detect parasites in sashimi. Using this strategy, the detection accuracy of the validation set for Anisakis nematodes on the top and bottom of a sliced piece of sashimi were 91.67% and 82.14%, respectively. Thus, VIS/NIR hyperspectral imaging allows for intelligent, accurate, efficient, and rapid detection of Anisakis nematodes on sashimi.

Classification of Codling Moth-Infested Apples Using Sensor Data Fusion of Acoustic and Hyperspectral Features Coupled with Machine Learning

Codling moth (CM) is a major apple pest. Current manual method of detection is not very effective. The development of nondestructive monitoring and detection methods has the potential to reduce postharvest losses from CM infestation. Previous work from our group demonstrated the effectiveness of hyperspectral imaging (HSI) and acoustic methods as suitable techniques for nondestructive CM infestation detection and classification in apples. However, both have limitations that can be addressed by the strengths of the other. For example, acoustic methods are incapable of detecting external CM symptoms but can determine internal pest activities and morphological damage, whereas HSI is only capable of detecting the changes and damage to apple surfaces and up to a few mm inward; it cannot detect live CM activity in apples. This study investigated the possibility of sensor data fusion from HSI and acoustic signals to improve the detection of CM infestation in apples. The time and frequency domain acoustic features were combined with the spectral features obtained from the HSI, and various classification models were applied. The results showed that sensor data fusion using selected combined features (mid-level) from the sensor data and three apple varieties gave a high classification rate in terms of performance and reduced the model complexity with an accuracy up to 94% using the AdaBoost classifier, when only six acoustic and six HSI features were applied. This result affirms that the sensor fusion technique can improve CM infestation detection in pome fruits such as apples.

Extended Depth of Field Imaging for Mosaic Hyperspectral Images.

Hyperspectral imaging (HSI) is an emerging modality for digital pathology. The purpose of this study is to develop an extended depth of field (EDOF) method for mosaic hyperspectral images acquired with a snapshot camera. EDOF is a technique for ensuring that an image is in focus at all points. A stack of mosaicked hyperspectral images of hematoxylin and eosin (H&E)-stained histologic slides were acquired at different positions along the z-axis and used to output a hyperspectral histologic image that was in-focus at every point. Three different methods were compared to achieve a fully focused image. We compared conventional patch-based methods to our proposed growth-based and band-based methods. The Brenner function was used to quantitatively measure the focus quality of each image measured. The results show that both of our proposed methods performed better qualitatively and quantitatively than the patch-based method, with the band-based method performing the best, as it leveraged dividing pixels into their proper wavelengths in addition to spatially, giving the algorithm better contrast to measure. In terms of speed, the band-based method was the fastest, followed by the patch-based method, with the growth-based method being the slowest. Our proposed extended depth of field hyperspectral imaging methods can have immediate applications in digital pathology, especially whole slide imaging, and other microscopic imaging.

Automated segmentation of foodborne bacteria from chicken rinse with hyperspectral microscope imaging and deep learning methods

Visible/near-infrared hyperspectral microscope imaging (HMI) has provided spectral-spatial features to identify pathogenic bacteria with high accuracy. But the bacterial detection with dark-field HMI requires accurate segmentation of single-cell bacteria from hyperspectral image (hypercube). In this study a robust technique was developed and evaluated to automatically segment single-cell pathogenic bacteria using deep learning and image processing. The proposed method consists of two steps as 1) bacterial segmentation with a deep learning model and 2) single-cell identification by ellipse fitting. Bacterial strains including Escherichia coli, Listeria, Salmonella, and Staphylococcus were prepared to obtain hyperspectral images of bacterial cells under different growth conditions with a Fabry–Perot Interferometer (FPI) HMI system. Based on the hypercube, four deep learning models including U-Net, residual U-Net (ResU-Net), attention gate residual U-Net (AGResU-Net), and attention-gated recurrent residual U-Net (AGR2U-Net) were employed for bacterial cell segmentation. AGR2U-Net with deblurred input images and −1 image padding performed better than other models with 94.1% mean intersection over union and visual inspection confirmed that segmented images with the model were identical to the ground-truth mask images. Also, ellipse fitting and goodness-of-fit evaluation were accurate in 97.4% of 6,426 examined cases. In addition, the robustness of the proposed method was confirmed because its segmentation accuracy and quality were moderately invariant with image blurriness and sample growth conditions. This accurate and robust auto-segmentation technique streamlined the detection of pathogenic bacteria with FPI-HMI by reducing processing time from raw hypercube acquisition to classification with 15 sec.

Single-Particle Hyperspectral Imaging for Monitoring of Gold Nanoparticle Aggregates in Macrophages.

Gold nanoparticles (Au NPs) have a promising future in the diagnosis and therapy of diseases owing to their unique photoelectric properties. In the body, monodisperse Au NPs may aggregate extracellularly and intracellularly, which influences their in vivo fate and physiological effects. However, due to the lack of a rapid, precise, and high throughput method for characterizing Au NP aggregates, the intricate aggregation process of Au NPs has not yet been fully understood. In order to overcome this obstacle, we develop a single-particle hyperspectral imaging method to identify Au NP aggregates, making use of the outstanding plasmonic properties of monodisperse and aggregated Au NPs. This method enables us to monitor the dynamic formation of Au NP aggregates in biological mediums and cells. Further single-particle hyperspectral imaging analysis reveals that following exposure to 100 nm Au NPs, the formation of Au NP aggregates in macrophages highly depends on exposure dosage and is less affected by exposure duration. In particular, surface coating, including PEGylation and protein corona, can considerably mitigate the intracellular aggregation of Au NPs. Our findings highlight that single-particle hyperspectral imaging is an efficient way for studying how Au NPs aggregate in biological scenarios.

A Comprehensive Correction Method for Radiation Distortion of Multi-Strip Airborne Hyperspectral Images

Airborne hyperspectral imaging plays an increasingly important role in environmental monitoring. However, due to the limitations of the acquisition conditions, there are uneven radiation and chromatic aberrations in the mosaic data. Accurate preprocessing of the original data is the premise of qualitative and quantitative remote sensing. In this study, we proposed a comprehensive radiation distortion correction method that integrates radiation attenuation difference correction, topographic correction, and multi-strip images consistency adjustment (RA-TOC-CA). First, the radiation attenuation equation was constructed by combining the viewing geometry, terrain, and the elevation difference between the UAV and the ground to eliminate the radiation attenuation difference of pixels acquired at the different instantaneous field of view (IFOV). Second, an improved kernel-driven BRDF model was built combining terrain information and illumination-viewing (flight attitude and sensor IFOV) geometry to eliminate the radiation unevenness and BRDF distortion caused by topography. Third, adjusting the reflectance of multi-strip images according to the homonymous points’ reflectance of adjacent strips should be equal, eliminating the radiation differences between multiple strips. Based on multi-strip airborne hyperspectral images collected in the Shaanxi province of China, the correction results of the RA-TOC-CA method were compared with those of the SCS+C and Minnaert+SCS methods regarding various evaluation criteria. The results showed that SCS+C and Minnaert+SCS can reduce the topographic effect but cannot eliminate the reflectance difference at the edges of adjacent images, and SCS+C overcorrects the reflectance. RA-TOC-CA weakened the topographic effects and brightness gradient, which was physically stable and generalizable. Compared with previous studies, RA-TOC-CA provided a complete radiation distortion correction method for airborne hyperspectral images and had a solid theoretical basis. This study introduces an effective method for radiation distortion correction of airborne hyperspectral images and provides technical support for large-scale applications of hyperspectral images.

MHealth hyperspectral learning for instantaneous spatiospectral imaging of hemodynamics.

Hyperspectral imaging acquires data in both the spatial and frequency domains to offer abundant physical or biological information. However, conventional hyperspectral imaging has intrinsic limitations of bulky instruments, slow data acquisition rate, and spatiospectral trade-off. Here we introduce hyperspectral learning for snapshot hyperspectral imaging in which sampled hyperspectral data in a small subarea are incorporated into a learning algorithm to recover the hypercube. Hyperspectral learning exploits the idea that a photograph is more than merely a picture and contains detailed spectral information. A small sampling of hyperspectral data enables spectrally informed learning to recover a hypercube from a red-green-blue (RGB) image without complete hyperspectral measurements. Hyperspectral learning is capable of recovering full spectroscopic resolution in the hypercube, comparable to high spectral resolutions of scientific spectrometers. Hyperspectral learning also enables ultrafast dynamic imaging, leveraging ultraslow video recording in an off-the-shelf smartphone, given that a video comprises a time series of multiple RGB images. To demonstrate its versatility, an experimental model of vascular development is used to extract hemodynamic parameters via statistical and deep learning approaches. Subsequently, the hemodynamics of peripheral microcirculation is assessed at an ultrafast temporal resolution up to a millisecond, using a conventional smartphone camera. This spectrally informed learning method is analogous to compressed sensing; however, it further allows for reliable hypercube recovery and key feature extractions with a transparent learning algorithm. This learning-powered snapshot hyperspectral imaging method yields high spectral and temporal resolutions and eliminates the spatiospectral trade-off, offering simple hardware requirements and potential applications of various machine learning techniques.

Evaluation of the effects of vacuum cooling on moisture contents, colour and texture of mushroom (Agaricus Bisporus) using hyperspectral imaging method

Visible near-infrared (400–1000 nm) hyperspectral imaging technology (HSI) was first time used to evaluate changes in various quality attributes including moisture contents (MC), lightness (L*), yellowness (b*), hardness (H) and chewiness (ch) of Agaricus bisporus as affected by vacuum cooling (VC). Partial least squares regression (PLSR) models based on Savitzky-Golay (SG) smoothing combined with the first derivative (SG-1st der) spectral preprocessing achieved good results for predicting MC with R2P of 0.823 and RMSEP of 0.341%. Similarly, PLSR models based on full wavelengths achieved good results for predicting L*, b* and H, and ch with R2P of 0.796, 0.848 and 0.810, and 0.903, with RMSEP of 1.253, 0.808 and 9.254, and 2.717. Finally, pseudocolour maps were developed to visualize the distribution of these attributes during VC. The current study provides a rapid, real-time and non-destructive detection strategy for evaluating the VC process.

A comparative study revealed hyperspectral imaging as a potential standardized tool for the analysis of cuticle tanning over insect development

Cereal-feeding beetles are a major risk for cereal crop maintenance. Cereal weevils such as Sitophilus oryzae have symbiotic intracellular bacteria that provide essential aromatic amino acid to the host for the biosynthesis of their cuticle building blocks. Their cuticle is an important protective barrier against biotic and abiotic stresses, providing high resistance from insecticides. Quantitative optical methods specialized in insect cuticle analysis exist, but their scope of use and the repeatability of the results remain limited. Here, we investigated the potential of Hyperspectral Imaging (HSI) as a standardized cuticle analysis tool. Based on HSI, we acquired time series of average reflectance profiles from 400 to 1000 nm from symbiotic (with bacteria) and aposymbiotic (without bacteria) cereal weevils S. oryzae exposed to different nutritional stresses. We assessed the phenotypic changes of weevils under different diets throughout their development and demonstrated the agreement of the results between the HSI method and the classically used Red-Green-Blue analysis. Then, we compared the use of both technologies in laboratory conditions and highlighted the assets of HSI to develop a simple, automated, and standardized analysis tool. This is the first study showing the reliability and feasibility of HSI for a standardized analysis of insect cuticle changes.

Rapid and Nondestructive Identification of Origin and Index Component Contents of Tiegun Yam Based on Hyperspectral Imaging and Chemometric Method

Tiegun yam is a typical food and medicine agricultural product, which has the effects of nourishing the kidney and benefitting the lungs. The quality and price of Tiegun yam are affected by its origin, and counterfeiting and adulteration are common. Therefore, it is necessary to establish a method to identify the origin and index component contents of Tiegun yam. Hyperspectral imaging combined with chemometrics was used, for the first time, to explore and implement the identification of origin and index component contents of Tiegun yam. The origin identification models were established by partial least squares-discriminant analysis (PLS-DA), support vector machine (SVM), and random forest (RF) using full wavelength and feature wavelength. Compared with other models, MSC-PLS-DA is the best model, and the accuracy of the training set and prediction set is 100% and 98.40%. Partial least squares regression (PLSR), random forest (RF), and support vector regression (SVR) models were used to predict the contents of starch, polysaccharide, and protein in Tiegun yam powder. The optimal residual predictive deviation (RPD) values of starch, polysaccharide, and protein prediction models selected in this study were 5.21, 3.21, and 2.94, respectively. The characteristic wavelength extracted by the successive projections algorithm (SPA) method can achieve similar results as the full-wavelength model. These results confirmed the application of hyperspectral imaging (HSI) in the identification of the origin and the rapid nondestructive prediction of starch, polysaccharide, and protein contents of Tiegun yam powder. Therefore, the HSI combined with the chemometric method was available for conveniently and accurately determining the origin and index component contents of Tiegun yam, which can expect to be an attractive alternative method for identifying the origin of other food.