Search

The demand for achromatic ultrathin flat lenses has become increasingly stringent, particularly for high‐performance imaging and display applications. Despite significant progress in achromatic metasurface and diffraction lenses, no single material has yet been capable of constructing ultrathin achromatic flat lenses covering ultrabroad wavebands, including the visible and near‐infrared (NIR), due to the limitations of material bandgaps. This limitation complicates fabrication processes, integration, and miniaturization, often leading to instability. In this paper, making use of the dispersionless nature of graphene, high numerical aperture multi‐wavelength achromatic metalenses (MAGLs) made entirely from graphene is proposed and demonstrated. This approach, based on a partial intensity resonance (PIR) mechanism, requires no iterative algorithms. Two MAGLs for visible and communication bands, respectively, are designed and fabricated. Remarkably, the measured focal lengths only deviate by less than 0.15% from the desired values. The graphene metalens (GML) in the visible produced clear and high‐quality images of microscopic character and Brassica napus cells. The demonstrated MAGLs significantly simplify the fabrication process and enhance integration, miniaturization, and stability. Their unique single‐material design offers tremendous potential to replace conventional refractive lenses in applications such as virtual reality glasses, hyperspectral imaging systems, and fluorescence microscopes. [ABSTRACT FROM AUTHOR]

Spectral discrimination by animal visions such as human, bird, and butterfly is numerically compared by angular analysis of photo‐response (PR) which will be recorded at photo‐receptors. Hyperspectral imaging system is utilized to simulate various animal vision. Bird vision is acute to discriminate colors among different vegetables due to its evenly spaced and narrow spectral responsivity of photo‐receptors compared to that of human. Butterfly vision is excellent in discriminating red tomato ripening due to the exclusive photo‐receptors detecting only over 600 nm. Even real and fake fruits in the same perceived color for human is discriminated by bird vision. Artificial vision with finely resolved polychromatic artificial cone cells are demonstrated surpassing human vision using visible multispectral camera. This provides insights on designing novel bio‐inspired vision system. [ABSTRACT FROM AUTHOR]

Physical unclonable function (PUF) is attractive in modern encryption technologies. Addressing the disadvantage of slow data storage/authentication in optical PUF is paramount for practical applications but remains an on‐going challenge. Here, a highly efficient PUF strategy based on random structural color domains (SCDs) of cellulose nanocrystal (CNC) is proposed for the first time, combing with hyperspectral imaging system (HIS) for ultrafast storage and authentication. By controlling the growth and fusion behavior of the tactoids of CNC, the SCDs display an irregular and random distribution of colors, shapes, sizes, and reflectance spectra, which grant unique and inherent fingerprint‐like characteristics that are non‐duplicated. Based on images and spectra, these fingerprint features are used to develop two sets of PUF key generation methods, which can be respectively authenticated at the user‐end and the manufacturer‐front‐end that achieving a high coding capacity of at least 22304. Notably, the use of HIS greatly shortens the time of key reading and generation (≈5 s for recording, 0.5–0.7 s for authentication). This new optical PUF labels can not only solve slow data storage and complicated authentication in optical PUF, but also impulse the development of CNC in industrial applications by reducing color uniformity requirement. [ABSTRACT FROM AUTHOR]

Aims Main Methods Key Findings Significance To establish a system based on hyperspectral imaging and deep learning for the detection of cancer cells in metastatic lymph nodes.The continuous sections of metastatic lymph nodes from 45 oral squamous cell carcinoma (OSCC) patients were collected. An improved ResUNet algorithm was established for deep learning to analyze the spectral curve differences between cancer cells and lymphocytes, and that between tumor tissue and normal tissue.It was found that cancer cells, lymphocytes, and erythrocytes in the metastatic lymph nodes could be distinguished basing hyperspectral image, with overall accuracy (OA) as 87.30% and average accuracy (AA) as 85.46%. Cancerous area could be recognized by hyperspectral image and deep learning, and the average intersection over union (IOU) and accuracy were 0.6253 and 0.7692, respectively.This study indicated that deep learning‐based hyperspectral techniques can identify tumor tissue in OSCC metastatic lymph nodes, achieving high accuracy of pathological diagnosis, high work efficiency, and reducing work burden. But these are preliminary results limited to a small sample. [ABSTRACT FROM AUTHOR]

Crustose coralline algae (CCA) are a highly diverse group of habitat‐forming, calcifying red macroalgae (Rhodophyta) with unique adaptations to diverse irradiance regimes. A distinctive CCA phenotype adaptation, which allows them to maximize photosynthetic performance in low light, is their content of a specific group of light‐harvesting pigments called phycobilins. In this study, we assessed the potential of noninvasive hyperspectral imaging (HSI) in the visible spectrum (400–800 nm) to describe the phenotypic variability in phycobilin content of an Antarctic coralline, Tethysphytum antarcticum (Hapalidiales), from two distinct locations. We validated our measurements with pigment extractions and spectrophotometry analysis, in addition to DNA barcoding using the psbA marker. Targeted spectral indices were developed and correlated with phycobilin content using linear mixed models (R2 = 0.64–0.7). Once applied to the HSI, the models revealed the distinct phycoerythrin spatial distribution in the two site‐specific CCA phenotypes, with thin and thick crusts, respectively. This study advances the capabilities of hyperspectral imaging as a tool to quantitatively study CCA pigmentation in relation to their phenotypic plasticity, which can be applied in laboratory studies and potentially in situ surveys using underwater hyperspectral imaging systems. [ABSTRACT FROM AUTHOR]

The demand for identification of maize varieties has increased dramatically due to the phenomenon of mixed seeds and inferior varieties pretending to be high‐quality varieties continuing to occur. It is urgent to solve the problem of efficient and accurate identification of maize varieties. A hyperspectral image acquisition system was used to acquire images of maize seeds. Regions of interest (ROI) with an embryo size of 10 × 10 pixel were extracted, and the average spectral information in the range of 949.43–1709.49 nm was intercepted for the subsequent study in order to eliminate random noise at both ends. Savitzky–Golay (SG) smoothing algorithm and multiple scattering correction (MSC) were used to pretreat the full‐band spectrum. The feature wavelengths were screened by successive projection algorithms (SPA), competitive adaptive reweighted sampling (CARS) single screening, and two combinations of CARS‐SPA and CARS + SPA, respectively. Support vector machines (SVMs) and models optimized based on genetic algorithm (GA), particle swarm optimization (PSO) were established by using full bands (FB) and feature bands as the model input. The results showed that the MSC‐(CARS‐SPA)‐GA‐SVM model had the best performance with 93.00% of the test set accuracy, 8 feature variables, and a running time of 24.45 s. MSC pretreatment can effectively eliminate the scattering effect of spectral data, and the feature wavelengths extracted by CARS‐SPA can represent all wavelength information. The study proved that hyperspectral imaging combined with GA‐SVM can realize the identification of maize varieties, which provided a theoretical basis for maize variety classification and authenticity identification. [ABSTRACT FROM AUTHOR]

Background/purpose: Because atopic dermatitis (AD) is a chronic inflammatory skin condition that causes structural changes, there is a growing need for noninvasive research methods to evaluate this condition. Hyperspectral imaging (HSI) captures skin structure features by exploiting light wavelength variations in penetration depth. In this study, parameter‐based transfer learning was deployed to classify the severity of AD using HSI. Therefore, we aimed to obtain an optimal combination of classification results from the four models after constructing different source‐ and target‐domain datasets. Methods: We designated psoriasis, skin cancer, eczema, and AD datasets as the source datasets, and the set of images acquired via hyperspectral camera as the target dataset for wavelength‐specific AD classification. We compared the severity classification performances of 96 combinations of sources, models, and targets. Results: The highest classification performance of 83% was achieved when ResNet50 was trained on the augmented psoriasis dataset as the source, with the resulting parameters used to train the model on the target Near‐infrared radiation (NIR) dataset. The second highest classification accuracy of 81% was achieved when ResNet50 was trained on the unaugmented psoriasis dataset as the source, with the resulting parameters used to train the model on the target R dataset. ResNet50 demonstrated potential as a generalized model for both the source and target data, also confirming that the psoriasis dataset is an effective training resource. Conclusion: The present study not only demonstrates the feasibility of the severity classification of AD based on hyperspectral images, but also showcases combinations and research scalability for domain exploration. [ABSTRACT FROM AUTHOR]

A study published in the journal Skin Research & Technology examines the use of a 3D hyperspectral imaging system and convolutional neural networks (CNN) for the diagnosis of skin lesions. The study involved 175 lesions from 73 patients and found that the novel imaging system showed promising diagnostic performance. The researchers compared the outcomes of the CNN analysis to traditional imaging methods and found that the HSI-CNN system achieved higher specificities and did not require expertise from the user. The study suggests that the HSI-CNN system has potential for skin cancer diagnosis in primary healthcare settings and may be a useful tool for specialists in ambiguous cases. However, further studies and comparisons with other imaging modalities are needed to validate the system's performance in clinical practice. [Extracted from the article]

Impact damage is one of the key factors affecting the quality of honey peaches. Quantitative study of impact damage is of great significance for the sorting of postharvest quality of honey peaches. In order to realize the quantitative prediction of impact damage of honey peaches, the impact damage of honey peaches was quantitatively studied based on the fusion of color characteristics with reflection spectra (R), absorbance spectra (A), and Kubelka‐Munk spectra (K‐M). The mechanical parameters of honey peaches during collision were obtained using a single pendulum collision device. Reflectance spectra and color characteristics of damaged honey peaches were obtained by a hyperspectral imaging system. The R spectra were converted into A and K‐M spectra, and the partial least squares regression (PLSR) model was built based on the three spectra and the three spectra combined with color characteristics for quantitative prediction of mechanical parameters. The results show that the prediction performance of the PLSR model is improved by combining color features with spectral information. In order to eliminate the redundant information in the spectral data, the competitive adaptive reweighted sampling (CARS) algorithm was used to select the characteristic wavelengths of the three spectra, and the selected characteristic wavelengths were fused with the color features to establish the PLSR model. The results show that the PLSR model built by the characteristic wavelengths of the A spectrum combined with the color features has the best prediction performance for the mechanical parameters. The RP value for maximum force is 0.862, and the RP value for damage depth is 0.894. The results of this study not only provide the theoretical support for the quality sorting, packaging, and transportation of honey peaches but also provide the reference for the biomechanical properties of various agricultural products. In order to reduce impact damage during postharvest handling of honey peaches. In this study, the prediction model of mechanical parameters during honey peaches collision was built by using color features fused with reflectance, absorption, and K‐M spectra, respectively. The results show that spectral and image information fusion can improve the prediction performance of the model. The model built by fusing the color features with the characteristic wavelengths of absorption spectrum has the best performance. [ABSTRACT FROM AUTHOR]

Background: Lip investigations and characterizations in the literature are less prevalent than for skin, particularly on the topic of color diversity. However, as the consumer demand increases for a nude lip makeup result, that is, shades close to the bare lip color, the identification and modification of lip color is essential for the cosmetic industry. Objective: The objective was to highlight lip color diversity among three ethnicities (Caucasian, African and Hispanic), through the use of a spectral color measurement device especially adapted to the lip area, and to consider lip color ethnic specificities and overlaps. Materials and Methods: The inferior natural lip color was measured with a full‐face hyperspectral imaging system, SpectraFace (Newtone Technologies, Lyon, France), on 410 healthy women aged 19 to 68 (Caucasian French, Caucasian American, African American, and Hispanic American women). A hierarchical ascending classification, was deployed to determine clusters based on the lip colorimetric parameters along two strategies to identify the best statistical analysis to preserve the lip color diversity. Results: Lip color is a continuous color space, with great intra‐ethnic and inter‐ethnic diversity, especially for African American women in terms of chroma and lightness. Among the two strategies of data analysis, our two‐step statistical clustering analysis yielded 11 groups (i.e., 11 lip tones), revealing an accurate representation of the scope of diversity, but also of the overlaps. Conclusion: The 11 lip tones/colors could potentially serve as target shades for the development of a more diverse and inclusive range of lip cosmetics, such as nude lipsticks. [ABSTRACT FROM AUTHOR]

Assessing plant community traits is important for understanding how terrestrial ecosystems respond and adapt to global climate change. Field hyperspectral remote sensing is effective for quantitatively estimating vegetation properties in most terrestrial ecosystems, although it remains to be tested in areas with dwarf and sparse vegetation, such as the Tibetan Plateau. We measured canopy reflectance in the Tibetan Plateau using a handheld imaging spectrometer and conducted plant community investigations along an alpine grassland transect. We estimated community structural and functional traits, as well as community function based on a field survey and laboratory analysis using 14 spectral vegetation indices (VIs) derived from hyperspectral images. We quantified the contributions of environmental drivers, VIs, and community traits to community function by structural equation modelling (SEM). Univariate linear regression analysis showed that plant community traits are best predicted by the normalized difference vegetation index, enhanced vegetation index, and simple ratio. Structural equation modelling showed that VIs and community traits positively affected community function, whereas environmental drivers and specific leaf area had the opposite effect. Additionally, VIs integrated with environmental drivers were indirectly linked to community function by characterizing the variations in community structural and functional traits. This study demonstrates that community‐level spectral reflectance will help scale plant trait information measured at the leaf level to larger‐scale ecological processes. Field imaging spectroscopy represents a promising tool to predict the responses of alpine grassland communities to climate change. [ABSTRACT FROM AUTHOR]

Monitoring the moisture content is crucial for red ginseng quality control. In this study, we established a moisture content prediction model using partial least squares regression (PLSR) based on short‐wave infrared (SWIR) data of different ginseng types (whole and sliced) at various drying temperatures and times. The model parameters, including the R2 values, ratio of prediction deviation, and range error ratio, showed that the developed model was robust. Sliced ginseng showed a constant rate period, followed by a falling rate period, whereas whole ginseng showed only a falling rate period. The falling‐rate period coincided with the formation of a dry zone on the sample surface in the moisture distribution map, indicating that monitoring the moisture distribution map can predict the onset of the falling‐rate period. These results suggest that the SWIR technique is a promising nondestructive method for evaluating the drying process of red ginseng. Practical applications: Monitoring the moisture content of red ginseng is essential to prevent microbial spoilage during transportation. Traditional monitoring of water content is destructive and time‐consuming. Therefore, using hyperspectral imaging, we aimed to develop an efficient moisture prediction model for the drying process of red ginseng. We believe that our study makes a significant contribution to non‐destructive and rapid quality evaluation in red ginseng production. [ABSTRACT FROM AUTHOR]

The extensive applications of spectrum analysis across various fields have rendered the traditional desktop spectrometers unable to meet the market demand for portability and instantaneity. Reducing the size of spectrometers has become a topic of interest. Based on this trend, a novel type of computational spectrometer is developed and has been widely studied owing to its unique features. Such spectrometers do not need to integrate complex mechanical or optical structures, and most of them can achieve spectrum analysis by the properties of the material itself combines with the reconstruction algorithm. Impressively, a single‐detector computational spectrometer has recently been successfully realized based on in situ modulation of material properties. This not only enables the further miniaturization of the device, but also means that the footprint‐resolution limitation which has always existed in the field of hyperspectral imaging has been broken, opening a new era of image analysis. This review summarizes the classifications and principles of various spectrometers, compares the spectrum resolution performances of different types of spectrometers, and highlights the progress of computational spectrometers, especially the revolutionary single‐detector spectrometer. It is expected that this review will provide a positive impact on expanding the boundary of spectrum analysis and move hyperspectral imaging forward. [ABSTRACT FROM AUTHOR]

The mass production of polymers has forced forensic practitioners to reconsider traditional methods of traces analysis. The field of impression reconstruction, specifically markings in firearms, is a prime example. This overview offers a critical evaluation of the relevant published techniques for the reconstruction of serial number in polymers, which include destructive methods such as swelling, heat treatments and relief polishing as well as methods allowing for trace preservation, such as hyperspectral Raman imaging combined with multivariate statistical analysis for enhanced pertinent data extraction. It therefore provides a complementary compilation to existing protocols for metal substrates. The novelty of this work lies within its approach, specifically by establishing not only the mechanistic scientific explanation for suitable comprehension and application of the techniques, but also by properly assessing their relevance considering the use in a forensic science context. The potential of wide‐field imaging techniques, mainly auto‐fluorescence analysis, is suggested for faster acquisition and reduced data processing (i.e., decreased time and greater accessibility). Additionally, application of the acquired knowledge to other relevant forensic traces, such as failure analysis of 3D printed objects, is proposed. Emphasis is also placed on the relevance of a purposeful interpretative framework necessary to reconstruct the singular past of the obliterated serial number thus leading to the identification of the given object in which it is affixed. This article is categorized under:Forensic Chemistry and Trace Evidence > Fingermarks and Other MarksForensic Chemistry and Trace Evidence > Emerging Technologies and MethodsForensic Chemistry and Trace Evidence > Trace Evidence [ABSTRACT FROM AUTHOR]

Flexible optoelectronics is a rapidly growing field, with a wide range of potential applications. From wearable sensors to bendable solar cells, curved displays, and curved focal plane arrays, the possibilities are endless. The criticality of flexible photodetectors for many of these applications is acknowledged, however, devices that are demonstrated thus far are limited in their spectral range. In this study, flexible photodetectors are demonstrated using a VOx nanoparticle ink, with an extremely broad operating wavelength range of 0.4 to 20 µm. This ink is synthesized using a simple and scalable wet‐chemical process. These photodetectors operate at room temperature and exhibit minimal variance in performance even when bent at angles of up to 100 ° at a bend radius of 6.4 mm. In addition, rigorous strain testing of 100 bend and release cycles revealed a photoresponse with a standard deviation of only 0.55%. This combination of mechanical flexibility, wide spectral response, and ease of fabrication makes these devices highly desirable for a wide range of applications, including low‐cost wearable sensors and hyperspectral imaging systems. [ABSTRACT FROM AUTHOR]

Background: Grape composition is of high interest for producing quality wines. For that, grape analyses are necessary, and require sample preparation, whether with classical analyses or with NIR analyses. The aim of the study was to test the ability of hyperspectral imaging (HSI), a nondestructive analysis to assess their composition. For that, seven grape varieties were analyzed for two vintages. Partial least squares (PLS) and discriminate (PLS‐DA) and PLS‐R were realized respectively in order to classify the berries, to validate the data sets, and to provide models to assess grape composition after a 1st derivative data pretreatment. Results: HSI allowed a 100% good classification of the grape varieties. It showed good results to assess technological ripening parameters (sugar and acid contents) as well as phenolic content (TPI, Total Phenolics, Total Anthocyanins, Total Flavonoids, and their extractable equivalents) (globally R2 > 0.81). However, it was not possible to reach the color intensity of grapes. Conclusion: HSI led to generate good models to assess wine grape composition. The quality of the generated models was dependent on the color of grapes and the parameter considered. [ABSTRACT FROM AUTHOR]

The detection of corrosion, especially in early stages, is a key factor for cost reduction in the maintenance of steel infrastructure. However, manual inspection is time consuming and takes considerable effort of people and equipment. Remotely operated vehicles with application‐specific sensors may overcome this problem. Novel sensing approaches like hyperspectral imaging (HSI) systems in combination with machine learning algorithms open new pathways for the rapid inspection of large surface areas in complex environments. In contrast to conventional RGB imaging, HSI contains both spatial and spectral reflectivity information over the complete visual spectral range and near infrared. This offers improved material characterization and classification possibilities using the chemical properties contained in the local reflection spectrum. This paper presents advantages and disadvantages of hyperspectral imaging systems for the detection of corrosion of steel infrastructure as well as suitable wavelength ranges for use with remotely operated vehicles. Furthermore, in‐field measurements with HSI for the inspection of corrosion on steel structures are shown. [ABSTRACT FROM AUTHOR]

Microscopic hyperspectral imaging technology is a potential non‐destructive and non‐contact method for colour measurement of micrometre‐sized textile fibres. However, specularity on the fibre surface can distort the accurate colour information and affect the accuracy of the colour measurement. This paper proposed a specular‐constrained sparse approximation (SCSA) for specular‐diffuse reflection separation from hyperspectral images of wool fibres. First, a specular prior map is generated based on the lightness dissimilarity. Then the SCSA model is used to decompose the processed hyperspectral image A into low‐rank data L, sparse specularity data S constrained by the specular prior map, sparse noise E, and Gaussian noise N. A non‐linear logistic sigmoid function and a sparse approximation of A – L – N to S are used to improve the performance of specularity removal during iterative optimization. The experimental results show that the proposed method significantly preserves diffuse reflectance and texture details in the specular highlight regions to obtain actual spectral reflectance and chromatic values from hyperspectral images of wool fibres. [ABSTRACT FROM AUTHOR]

Plasmonic nanostructures enable the strong localization of an electromagnetic field, which allows them to be used for enhancing Raman signals. This property has a range of applications in various fields, from detecting small quantities of substances to studying the interactions between compounds and metals. Although the general mechanism of Raman signal enhancement by metal nanoparticles is accepted, details related to the interaction of the nanoparticle with the molecule are still unexplained because of its complexity. A versatile, high‐resolution microscope system is required to investigate the spectroscopic properties of nanoparticles and the Raman signals enhanced by them. This study describes a multifunctional microscope system designed for comprehensive microspectroscopic analysis of plasmonic nanostructures functionalized with chemicals. The integrated system combines various excitation and detection modalities, enabling hyperspectral Raman mapping, Rayleigh (dark‐field) scattering hyperspectral mapping, and angular scattering and emission mapping in defocused mode. The flexibility in excitation wavenumber choice ensures optimal measurement efficiency and the ability to perform all measurements as a function of excitation. We demonstrate the system's capabilities using malachite green deposited on gold nanorods, showcasing the surface‐enhanced Raman scattering (SERS) excitation profiles and scattering directionality defocused imaging using the hyperspectral imaging of the angular distribution of Raman scattering technique. This integrated microscope system promises to advance the study of plasmonic nanostructures and their interactions with chemical compounds at the single‐particle level. The construction of a multimodal Raman microscopic system that enables hyperspectral imaging of the angular distribution of Raman scattering and dark‐field scattering, as well as Raman excitation profiles, is presented. The capabilities of the system are demonstrated by studying the surface‐enhanced Raman scattering of malachite green deposited on gold nanorods. [ABSTRACT FROM AUTHOR]

Introduction: The lips are of significant interest among the facial features for self‐perception and social interactions. Such perceptions could rely in part upon the contrast in color between the lips and the surrounding skin. In contrast to skin color, lip color ethnic diversity measurement provides a challenge and is less documented in the literature, most likely due primarily to challenges in measurement based upon the lip characteristics of size and contour. Accurate measurement and reproduction of lip color are essential to build strong cross‐sectional knowledge about lip color. Objective: This preliminary experiment of a large‐scale project on lip and skin color was designed to assess accuracy in lip color measurement and recommend instruments and protocols for good measurement practice. Methods: The study involved 19 healthy French Caucasian women. Their inferior and superior natural lip color was measured twice with three devices in order to extract the standard colorimetric parameters of lightness (L*), chroma (C*), and hue (h): the VISIA‐CR® (high‐resolution digital camera, Canfield Scientific Inc.); the spectrophotometer VS3200 (45°:0° spectrophotometer, MetaVue™ VS3200, X‐Rite Inc.); and the SpectraFace® (new hyperspectral imaging system, Newtone Technologies). The intercomparison and color distribution between the different instruments and the measurement repeatability of each one were assessed. Results: Our results showed a reliable and equivalent repeatability for the three tested instruments. Moreover, the three tools led to the same conclusion regarding lip color heterogeneity (lighter, redder, less yellow and more saturated inferior lip versus superior lip). Agreement between the SpectraFace® and the spectrophotometer measurements was higher, and the SpectraFace® allowed to identify three separated lip color typologies. Conclusion: In conclusion, the combination of imaging technology and hyperspectral measurement makes the SpectraFace® a useful tool to investigate lip color characterization and establish a large‐scale cross‐sectional knowledge study about lip color. [ABSTRACT FROM AUTHOR]

To understand surface biogeophysical processes, accurately evaluating the geographical and temporal fluctuations of soil moisture is crucial. It is well known that the surface soil moisture content (SMC) affects soil reflectance at all solar spectrum wavelengths. Therefore, future satellite missions, such as the NASA Surface Biology and Geology mission, will be essential for mapping and monitoring global soil moisture changes. Our study compares two widely used moisture retrieval models: the multilayer radiative transfer model of soil reflectance (MARMIT) and the soil water parametric (SWAP)‐Hapke model. We evaluated the SMC retrieval accuracy of these models using unmanned aerial systems (UAS) hyperspectral imagery and goniometer hyperspectral data. Laboratory analysis employed hyperspectral goniometer data of sediment samples from four locations reflecting diverse environments, while field validation used hyperspectral UAS imaging and coordinated ground truth collected in 2018 and 2019 from a barrier island beach at the Virginia Coast Reserve Long‐Term Ecological Research site. The (SWAP)‐Hapke model achieves comparable accuracy to MARMIT using laboratory hyperspectral data but is less accurate when applied to UAS hyperspectral imagery than the MARMIT model. We proposed a modified version of the (SWAP)‐Hapke model, which achieves better results than MARMIT when applied to laboratory spectral measurements; however, MARMIT's performance is still more accurate when applied to UAS imagery. These results are likely due to differences in the models' descriptions of multiply‐scattered light and MARMIT's more detailed description of air‐water interactions. Plain Language Summary: Understanding how soil moisture content (SMC) changes is crucial because it plays a significant role in many environmental processes. This study compares three models using hyperspectral data and imagery from the laboratory and a barrier island field setting to find the most effective method for mapping SMC over large areas with the larger goal of application to global scales using the future NASA Surface Biology and Geology mission hyperspectral imagery. We considered three physics‐based models: the multilayer radiative transfer model of soil reflectance (MARMIT) model, the SWAP‐Hapke model, and a modified version of the SWAP‐Hapke model. To better model retrieval of SMC in the landscape from imagery, we used a hyperspectral imaging system, which has a higher information capacity than traditional RGB and multispectral imagery, to collect imagery over field validation sites from an unmanned aerial system (UAS), while the laboratory hyperspectral data was collected from samples, acquired in four different locations, using a hyperspectral goniometer. Differences in underlying model assumptions favor the modified (SWAP)‐Hapke model for laboratory data where illumination is solely from a directional source, while the MARMIT performs best with the field UAS imagery where both the direct solar and diffuse skylight illuminate the soil. Key Points: We compared three soil moisture retrieval models, multilayer radiative transfer model of soil reflectance (MARMIT), soil water parametric (SWAP)‐Hapke, and a variant, using hyperspectral UAS imagery and goniometer dataBest models were: modified (SWAP)‐Hapke model (directional source/goniometer); MARMIT (UAS imagery/direct solar and diffuse skylight)Differences in model treatment of the equivalent water layer likely contribute to these different outcomes depending on illumination [ABSTRACT FROM AUTHOR]

In traditional two‐dimensional space object detection, it is believed that 'objects with similar shapes belong to the same type', making it difficult to distinguish true and false targets. In order to improve the efficiency of real object localization and detection, a specific target localization and authenticity recognition method (STLAR) based on joint spatial and spectral information is proposed. Firstly, the ground imaging spectrometer is utilized to obtain hyperspectral image data of the area to be measured. Then use a spatial object detection model to frame specific targets. Finally, compare the similarity between specific targets detected in space and prior spectral information. The experimental results successfully completed the task of target localization and recognition of authenticity. The proposed method can simultaneously solve the problems of spectral dimension detection or spatial dimension detection, providing new ideas for future rapid localization and recognition of specific targets. [ABSTRACT FROM AUTHOR]

To address the major challenges to obtain high spatial resolution in snapshot hyperspectral imaging, 3D printed glass lightguide array is developed to sample the intermediate image in high spatial resolution and redistribute the pixels in the output end to achieve high spectral resolution. Curved 3D printed lightguide array can significantly simplify the snapshot hyperspectral imaging system, achieve better imaging performance, and reduce the system complexity and cost. Two‐photon polymerization process to print glass lightguide array is developed, and the system performance with biological samples is demonstrated. This new snapshot technology will catalyze new hyperspectral imaging system development and open doors for new applications from UV to infrared. [ABSTRACT FROM AUTHOR]

Rapid drug detection is vital for combating drug abuse and inhibiting transmission, and the emergence of various novel psychoactive substances (NPS) has placed higher demands on field testing. To establish a rapid, nondestructive, and accurate analytical and classification method for testing novel psychoactive substances, shifted‐excitation Raman difference spectroscopy (SERDS) was used to detect 37 NPS including fentanyl, amphetamine, and synthetic cannabinoid under the experimental conditions of laser light source (785 and 785.5 nm). SERDS was also used in conjunction with machine learning to find the best classification prediction model. In this study, the characteristic peaks of 37 NPS analogs were extracted, and the characteristic peaks were attributed to the structures of the substances. We compared the classification effects of Support Vector Machine (SVM), K‐NearestNeighbor (KNN), ensemble classifier, neural network, tree, naive Bayes, and discriminant analysis and provided three hyperparametric optimization methods. Finally, the cross‐validation accuracy of SVM under BOA optimization is 97.3%, which can well distinguish three families of samples. The ability of SERDS to resist strong fluorescence interference makes it a powerful tool for on‐site analysis of NPS. In this study, we have given 37 unreported standard Raman peaks of NPS and three seized samples, which can be used as a reference. Shifted‐excitation differential Raman spectroscopy combined with machine learning can effectively provide solutions for customs, health care, on‐site police, large event security, and trace evidence inspection. [ABSTRACT FROM AUTHOR]

Measuring skin melanin concentration in order to assess skin phototype according to Fitzpatrick's classification is a constant research goal. In this study, a new approach for assessing skin melanin concentration based on hyperspectral imaging combined with an appropriate analytical model that exploits specific spectral bands to generate maps of melanin content distribution on different Fitzpatrick skin phototypes is presented. Hyperspectral images from the proximal inner side of the forearms of 51 young volunteers covering the first four classes of Fitzpatrick's phototypes were acquired using a hyperspectral imaging system. The images were analyzed using a modified Beer–Lambert law that segregates the contribution of melanin from the other constituents to the skin absorption spectrum. The performance of the model was evaluated using the coefficient of determination (r‐squared). The results revealed that the approach proposed in this study generated accurate melanin concentration distribution maps that allowed a correct classification of skin phototype. In conclusion, the proposed approach for assessing skin melanin concentration proved to be very reliable for classifying skin phototypes, and, as it provides maps that are easily read, it has the advantage of a possible extension of its applications to other research concerning skin pigmentation. [ABSTRACT FROM AUTHOR]

Large and publicly available soil spectral libraries, such as the USDA National Soil Survey Center–Kellogg Soil Survey Laboratory (NSSC‐KSSL) mid‐infrared (MIR) spectral library, are enormously valuable resources enabling laboratories around the world to make rapid low‐cost estimates of a number of soil properties. A limitation to widespread sharing of soil spectral data is the need to ensure that spectra collected on a secondary spectrometer are compatible with the spectra in the primary or reference library. Various spectral preprocessing and calibration transfer techniques have been proposed to overcome this limitation. We tested the transferability of models developed using the USDA NSSC‐KSSL MIR library to a secondary instrument. For the soil properties, total carbon (TC), pH, and clay content, we found that good performance (ratio of performance to deviation [RPD] = 4.9, 2.0, and 3.6, respectively) could be achieved on an independent test set with Savitzky‐Golay smoothing and first derivative preprocessing of the secondary spectra using a memory‐based learning chemometric approach. We tested three calibration transfer techniques (direct standardization [DS], piecewise direct standardization [PDS], and spectral space transformation [SST]) using different size transfer sets selected to be representative of the entire NSSC‐KSSL library. Among the transfer methods, SST consistently outperformed DS and PDS with 50 transfer samples being an optimal number for transfer model development. For TC and pH, performance was improved using the SST transfer (RPD = 7.7 and 2.2, respectively) primarily through the elimination of bias. Calibration transfer could not improve predictions for clay. These findings suggest that calibration transfer may not always be necessary, but users should test to confirm this assumption using a small set of representative samples scanned at both laboratories. CORE IDEAS: Spectroscopy is becoming a common tool in soil science.Sharing spectral libraries is still a challenge due to compatibility issues.Spectral pretreatment helped standardize spectra and produced good predictions.Calibration transfer could improve performance for some properties by eliminating bias. [ABSTRACT FROM AUTHOR]

In order to the quickly and nondestructively detect whether starch is adulterated in minced chicken meat, a novel method combining hyperspectral imaging (HSI) technique with transfer learning was proposed in this study. First of all, hyperspectral images of minced chicken meat with different mass fractions of starch were collected and spectral information of the samples in the range of 400.89–1000.19 nm was extracted. Then, the hyperspectral data was preprocessed via continuous wavelet transform (CWT), which transformed the pixel‐level hyperspectral data into two‐dimensional spectrograms. Furthermore, classification model for identifying starch in minced chicken meat was constructed based on the GoogLeNet network pretrained on ImageNet data set. Finally, the support vector machine (SVM) model and the convolutional neural network (CNN) model without transfer learning were established for comparison. The results indicated that the model based on GoogLeNet network had a higher classification accuracy, up to 98.6%. Therefore, this study demonstrates the feasibility of the detection of starch in minced chicken meat based on HSI technique and transfer learning. Practical applications: The identification of starch in the minced chicken meat is of particular significance for maintaining the market order and safeguarding the human health. An innovative method based on hyperspectral imaging technique and transfer learning to identify whether the minced chicken meat mixed with starch was proposed in this study. The method achieved quickly, nondestructively and accurately detection of starch in minced chicken meat. [ABSTRACT FROM AUTHOR]

Snapshot hyperspectral imaging technology is increasingly used in agricultural product monitoring. In this study, we present a 9× local zoom snapshot hyperspectral imaging system. Using commercial spectral sensors with spectrally resolved detector arrays, we achieved snapshot hyperspectral imaging with 14 wavelength bands and a spectral bandwidth of 10–15 nm. An experimental demonstration was performed by acquiring spatial and spectral information about the fruit and Drosophila. The results show that the system can identify Drosophila and distinguish well between different types of fruits. The results of this study have great potential for online fruit classification and pest identification. [ABSTRACT FROM AUTHOR]

The feasibility of applying a low‐cost and portable NIR spectrometer to detect the S‐ovalbumin content of eggs was investigated in this study, and a model population analysis based competitive adaptive reweighted sampling (MPA‐CARS) algorithm was proposed to reduce the dimensionality of spectral data. The NIR spectra of egg samples with different storage periods were collected in the wavelength range of 900–1700 nm. Standard normal variate (SNV) and Savitzky–Golay1st derivative were used to preprocess the raw spectral data. Then, CARS and the proposed MPA‐CARS were applied to select effective wavelengths from the whole spectral range. Statistical results showed that MPA‐CARS had better feature extraction performance than CARS and the number of selected feature wavelengths was less. Support vector machine (SVM), back propagation neural network (BPNN) and extreme gradient boosting (XGBoost) were used to establish calibration models for predicting S‐ovalbumin content, in which the simplified XGBoost model based on MPA‐CARS feature wavelengths yielded the best performance, with R2P of 0.906 and RMSEP of 7.799%. Therefore, portable NIR has the potential to be a useful tool for S‐ovalbumin content detection. This could help food processing industry to arrange miniaturized NIR sensors to detect egg quality at different points in the egg supply chain. Practical applications: S‐ovalbumin content is a freshness index of eggs and can affect the quality of processed foods. This study demonstrated the feasibility of using portable NIR spectrometer to detect the S‐ovalbumin content of eggs, which can help food processing agencies to implement low‐cost NIR sensors to detect egg quality at different stages of the supply chain. [ABSTRACT FROM AUTHOR]

Although hyperspectral data, especially spaceborne images, are rich in spectral information, their spatial resolution is usually low due to the limitation of sensor design and other factors. Therefore, for the application of hyperspectral images, unmixing technology is a key processing technology, such as linear mixing model and its derived algorithms have made a certain progress. However, a real scene often contains both pure and mixed pixels. The existing methods usually ignore the consideration and analysis of this situation in the process of model design and simulation experiment. In this context, this paper proposes a robust linear unmixing model with the enhanced constraint of classification for hyperspectral image. In general, it designs a framework combining unmixing and classification. In the task for real scene data, endmembers are extracted first, and then the hard classification term constructed after the expansion of endmembers (training samples) based on similarity is introduced to provide the sparsity constraint of the overall model, so as to realize relatively complete adjustment and effective image unmixing under complex conditions. Considering the scene with different distributions, the simulation experiment designs several groups of data tests, including different proportions of pure and mixing pixels. The unmixing results of three simulated datasets and two real datasets show that the unmixing results of this method are better than those of the other six comparison methods. This model improves the accuracy of unmixing and realizes effective unmixing. [ABSTRACT FROM AUTHOR]

Hyperspectral data are gaining popularity in remote sensing and signal processing communities because of the increased spectral information relative to multispectral data. Several airborne and spaceborne hyperspectral datasets are publicly available, facilitating the development of various applications and algorithms. However, hyperspectral data are usually limited by their narrow, highly correlated and contiguous spectral bands in both processing and analysis. Moreover, the resolution of available hyperspectral datasets is not sufficiently high for the identification of small objects. Nevertheless, with the rapidly advancing technology, hyperspectral imaging systems can now be mounted on small aerial vehicles for detecting small objects at low altitude. To properly handle these high spectral and spatial resolution data, new or redesigned data processing or analysis pipelines must be developed, but such datasets are limited. In this study, we describe two hyperspectral datasets acquired by a drone and evaluate their radiometric and geometric quality. Based on appropriate data acquisition and processing approaches, our datasets are expected to be useful as testbeds for new algorithms and applications. [ABSTRACT FROM AUTHOR]

Prior knowledge of textile fiber colors in blends is useful for color recipe assessment. There have been various methods to improve the accuracy of color recognition from the dataset of colored textile fibers in recent years. However, numerical assessments based on spectral feature and color difference is insufficient, in which the accuracy of color recognition can be affected by morphology and the uneven coloration on a single fiber. This paper proposes a novel 3D convolutional neural network model (3D‐CNN) with supervised spectral regression for the color recognition of hyperspectral images (HSI) of colored textile fiber. The proposed method obtained spatial‐spectral features based on 3D‐CNN, and the true spectrum of each class was used for supervised spectral regression to improve the accuracy. The loss function used was the sum of the supervised classification loss function and the spectral regression loss function model are optimized by mini‐batch‐based backpropagation. The proposed method was trained and tested on the HSI dataset composed of 100 colors of wool fibers acquired through a microscopic hyperspectral imaging system at a ×3.375 optical magnification. The experimental results showed that the proposed method exhibited better performance compared to numerical assessments and other deep learning models, except for efficiency. Specifically, it achieved better recognition performance on sub‐datasets of similar colored and light‐colored wool fiber where subtle inter‐class and large intra‐class variance existed. [ABSTRACT FROM AUTHOR]

Illegal merchants adulterate the powder of panax notoginseng rhizome and main root with the powder of panax notoginseng fibrous root, which reduces the efficacy and hygiene of panax notoginseng powder. To detect the adulteration of panax notoginseng powder nondestructively and rapidly, in this study, the hyperspectral images of 80 samples of adulterated rhizome powder and adulterated main root powder (AR & AM) were collected by using a hyperspectral image acquisition system (400–1000 nm). Then, Savitzky–Golay and standard normalized variable (SG‐SNV) were used to preprocess the spectrum and improve the signal‐to‐noise ratio of the data. Next, competitive adaptive reweighed sampling (CARS) and iteratively retains informative variables (IRIV) methods were respectively adopted to select feature wavelengths from the spectra data after pretreatment. Subsequently, support vector regression (SVR) was used to establish the prediction models based on the selected variables and the full spectra data. Also, the arithmetic optimization algorithm (AOA) was used to optimize the parameters of the SVR model. The results showed that CARS‐AOA‐SVR was the optimal model for AR & AM, achieving RP2 of.9693 and.9667, RMSEP of.0266 and.0264, respectively. Therefore, the hyperspectral imaging combined with the CARS‐AOA‐SVR is a feasible method to detect the adulteration concentration of panax notoginseng powder. Practical Applications: The traditional detection methods for adulteration of panax notoginseng powder depended on manual detection and chemical analysis. In order to detect the adulterated concentration of panax notoginseng powder rapidly and accurately, the adulterated panax notoginseng powder was studied based on hyperspectral imaging. The results showed that hyperspectral imaging combined with CARS‐AOA‐SVR model could be used to quantitatively detect the adulteration concentration of panax notoginseng powder, which could be a fast, accurate, and nondestructive method for adulteration detection of panax notoginseng powder and other powdery materials. [ABSTRACT FROM AUTHOR]

Deep learning approaches have significantly enhanced the classification accuracy of hyperspectral images (HSIs). However, the classification process still faces difficulties such as those posed by high data dimensions, large data volumes, and insufficient numbers of labeled samples. To enhance the classification accuracy and reduce the data dimensions and training needed for labeled samples, a 3D fully convolutional neural network (3D-FCNN) model was developed by including a bottleneck attention module. In such a model, the convolutional layer replaces the downsampling layer and the fully connected layer, and 3D full convolution is adopted to replace the commonly used 2D and 1D convolution operations. Thus, the loss of data in the dimensionality reduction process is effectively avoided. The bottleneck attention mechanism is introduced in the FCNN to reduce the redundancy of information and the number of labeled samples. The proposed method was compared to some advanced HSI classification approaches with deep networks, and five common HSI datasets were employed. The experiments showed that our network could achieve considerable classification accuracies by reducing the data dimensionality using a small number of labeled samples, thereby demonstrating its potential merits in the HSI classification process. [ABSTRACT FROM AUTHOR]

Automated recognition of road surface objects is vital for efficient and reliable road condition assessment. Despite recent advances in developing computer vision algorithms, it is still challenging to analyze road images due to the low contrast, background noises, object diversity, and variety of lighting conditions. Motivated by the need for an improved pavement objects classification, we present Dual Attention Convolutional Neural Network (DACNN) to improve the performance of multiclass classification using intensity and range images collected with 3D laser imaging devices. DACNN fuses heterogeneous information in intensity and range images to enhance distinguishing foreground from background, as well as to improve object classification in noisy images under various illumination conditions. DACNN also leverages multiscale input images by capturing contextual information for object classification with different sizes and shapes. DACNN contains an attention mechanism that (i) considers semantic interdependencies in spatial and channel dimensions and (ii) adaptively fuses scale-specific and mode-specific features so that each feature has its own level of contribution to the final decision. As a practical engineering project, dataset are collected from road surfaces using 3D laser imaging. DACNN is compared with four deep classifiers that are widely used in transportation applications. Experiments show that DACNN consistently outperforms the baselines by 22–35% on average in terms of the F-score. A comprehensive discussion is also presented regarding computational costs and how robustly the investigated classifiers perform on each road object. [ABSTRACT FROM AUTHOR]

Trace inspection is a key technology for collecting crime scenes in the criminal investigation department. A lot of information can be obtained by restoring and analyzing the remaining traces on the scene. However, with the development of digital technology, digital trace inspection has become more and more popular. So, the main research of this article is the design and realization of the trace inspection system based on hyperspectral imaging technology. This article proposes nondestructive testing technology in hyperspectral imaging technology. Combining basic principles of spectroscopy and the image of residual traces such as car tires, shoe soles, and blood stains, it can identify the key traces. Then, based on the image denoising and least squares support vector machine method, this study improves the accuracy and restoration of the image. Therefore, this study designs a test for the trace inspection system for testing hyperspectral imaging technology. The test items include the performance of the trace inspection system, the noise reduction of the trace inspection system, and the ability of the trace inspection system to inspect blood stains. The final collected data are improved to get the trace inspection system based on hyperspectral imaging technology proposed in this study. Compared with the traditional trace inspection system, the experimental results show that the trace inspection system based on hyperspectral imaging technology can improve the accuracy by 5%–28%, compared with the traditional trace inspection system. The image restoration degree of the hyperspectral imaging technology trace inspection system can be improved by 1%–19%, compared with the traditional trace inspection system. [ABSTRACT FROM AUTHOR]

In order to improve the effectiveness of visual communication design, this article combines hyperspectral image processing technology to construct a visual communication design system and improves the effect of visual communication design with intelligent methods. Eagerly, this article proposes a general framework for constructing a suitable shock capture scheme in combination with the entropy control method for suppressing shock instability. In addition, considering the guarantee of qualitative and entropy stability, this article adopts a variety of numerical experiments to evaluate the performance of the full-speed domain scheme. The simulation study shows that the visual communication system based on hyperspectral image processing technology proposed in this article has a very good visual communication effect and can effectively improve the visual communication design effect. [ABSTRACT FROM AUTHOR]

Background: Color imaging is a tried and true method for the evaluation of cosmetic and dermatological effects, but it fails to capture all the information in a scene's spectral reflectance. For this reason, there has been in recent years increasing interest in the use of imaging spectrometers for clinical studies and product evaluation. Material and methods: We developed a novel HyperSpectral Imager (HSI) able to take in vivo full‐face format images as a next generation instrument for skin color measurement and beyond. Here, we report part of the results of our first full‐scale validation test of the HSI. We replicated a make‐up foundation screening test by applying three products to a panel of 9 models and evaluated the product L∗, a∗, b∗, and ∆E effect immediately after application relative to the bare skin condition. We repeated this test twice in order to study the repeatability of the HSI as an evaluation instrument and during each test two different operators duplicated the data acquisition so we can assess the reproducibility of the measurements. Results: We find that the measurements from the HSI provide repeatability and reproducibility as good or better than those of our previous benchmark devices. Conclusion: From these results, we conclude that not only is the HSI suitable for use in color evaluation studies, but also that it gives operational advantages over the previous generation of evaluation instruments, as it provides a spectral measurement combined with good spatial resolution. This allows for analysis of color over an area and post hoc selection of study regions and so opens new possibilities for studies of complex in vivo phenomena which neither non‐imaging spectrometers nor conventional cameras can pursue. This study also raises points for future work concerning proper inclusion of instrument uncertainty in comparisons of results between instruments and handling of systematic uncertainties from analyses based on a single area. [ABSTRACT FROM AUTHOR]

Summary: Panax notoginseng saponin (PNS) is the most important physical and chemical index of panax notoginseng. In order to detect PNS rapidly and non‐destructively, 160 hyperspectral images of panax notoginseng rhizome and main root were acquired by using a visible‐near infrared hyperspectral image acquisition system (400–1000 nm), and the original spectrum were extracted from hyperspectral images. The signal‐to‐noise ratio of the spectrum was improved by savitzky‐golay mixed multiplication scatter correction (SG‐MSC) pretreatment. Feature wavelengths were extracted by using competitive adaptive reweighted sampling (CARS), variable combination population analysis (VCPA) and bootstrapping soft shrinkage (BOSS), and support vector regression (SVR) model was established based on the feature spectrum and the original spectrum. By comparing, it was found that BOSS had the best effect of feature selection. In order to improve the accuracy of the model, equilibrium optimizer (EO) was used to optimise the parameters (c, g) of the BOSS‐SVR model. The results showed that BOSS‐EO‐SVR of the optimal prediction model of PNS, achieving RP2 and RMSEP of 0.95 and 0.32%, respectively. Therefore, hyperspectral imaging combined with BOSS‐EO‐SVR model is a feasible method to detect PNS. [ABSTRACT FROM AUTHOR]

Compressed hyperspectral imaging is a powerful technique for satellite-borne and airborne sensors that can effectively shift the complex computational burden from the resource-constrained encoding side to a presumably more capable base-station decoder. Reconstruction algorithms play a pivotal role in compressive imaging systems. Traditional model-based reconstruction approaches are computationally burdensome and achieve limited success. Deep learning-based approaches, while improving in reconstruction accuracy and speed, depend heavily on data, which is a major challenge for satellite-borne hyperspectral compressed imaging. In this article, we combine the respective advantages of model-based and learning-based approaches in a distributed compressed hyperspectral sensing framework, employing linear mixed model assumptions and spectral library learning to simultaneously improve the reconstruction speed and accuracy without using a large amount of additional hyperspectral data. First, the relationship between the CS band and the key band is learned from the spectral library to ensure that the key band endmembers can be accurately predicted. Then, the joint horizontal and vertical difference operators are proposed to enhance the estimation of the initial values of abundance. Finally, the CS band endmembers and residuals are updated in the reconstruction module to deal with the endmember and abundance mismatch. Extensive experimental results on five real hyperspectral datasets demonstrate that the proposed spectral library learning, abundance initialization, and reconstruction strategy can effectively improve the compressed sensing reconstruction accuracy, outperforming the existing state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Magnetic resonance imaging (MRI) is an accurate and noninvasive method employed for the diagnosis of various kinds of diseases in medical imaging. Most of the existing systems showed significant performances on small MRI datasets, while their performances decrease against large MRI datasets. Hence, the goal was to design an efficient and robust classification system that sustains a high recognition rate against large MRI dataset. Accordingly, in this study, we have proposed the usage of a novel feature extraction technique that has the ability to extract and select the prominent feature from MRI image. The proposed algorithm selects the best features from the MRI images of various diseases. Further, this approach discriminates various classes based on recursive values such as partial Z-value. The proposed approach only extracts a minor feature set through, respectively, forward and backward recursion models. The most interrelated features are nominated in the forward regression model that depends on the values of partial Z-test, while the minimum interrelated features are diminished from the corresponding feature space under the presence of the backward model. In both cases, the values of Z-test are estimated through the defined labels of the diseases. The proposed model is efficiently looking the localized features, which is one of the benefits of this method. After extracting and selecting the best features, the model is trained by utilizing support vector machine (SVM) to provide the predicted labels to the corresponding MRI images. To show the significance of the proposed model, we utilized a publicly available standard dataset such as Harvard Medical School and Open Access Series of Imaging Studies (OASIS), which contains 24 various brain diseases including normal. The proposed approach achieved the best classification accuracy against existing state-of-the-art systems. [ABSTRACT FROM AUTHOR]

The recent progress of machine learning and microfluidics in the chemical and biological sciences has motivated the development of new online techniques to interrogate the (bio)chemical contents within moving droplets. To accelerate the optical characterization of new materials and chemical reactions, we combine a line‐scan hyperspectral imaging system with a droplet‐based microfluidic reactor. We demonstrate the performance of this platform on a model chemistry – silver nanoparticle synthesis. The platform can image the spectral signature of ∼400 individual droplets in only 15 s, with droplet flow speeds exceeding 4 cm/s in the reaction tube. After correction of the keystone and smile effects on the hyperspectral images, the absorbance spectra are extracted from the droplets with an accuracy comparable to industrial spectrophotometers. The time evolution of the UV/Vis absorbance spectra during the reactive synthesis can be tracked either by scanning all the droplets present in the reaction tube or by following a subset of the droplet ensemble at frame rates up to 92 fps. This high‐throughput and high‐speed platform is particularly interesting for screening large parameter spaces and imaging fast reactions with a high resolution, for eventual coupling with advanced machine learning techniques to infer kinetic models and obtain detailed mechanistic insights. [ABSTRACT FROM AUTHOR]

In this paper, an in-depth study of cross-media semantic matching and user adaptive satisfaction analysis model is carried out based on the convolutional neural network. Based on the existing convolutional neural network, this paper uses rich information. The spatial correlation of cross-media semantic matching further improves the classification accuracy of hyperspectral images and reduces the classification time under user adaptive satisfaction complexity. Aiming at the problem that it is difficult for the current hyperspectral image classification method based on convolutional neural network to capture the spatial pose characteristics of objects, the problem is that principal component analysis ignores some vital information when retaining a few components. This paper proposes a polymorphism based on extension Attribute Profile Feature (EMAP) Stereo Capsule Network Model for Hyperspectral Image Classification. To ensure the model has good generalization performance, a new remote sensing image Pan sharpening algorithm based on convolutional neural network is proposed, which increases the model's width to extract the feature information of the image and uses dilated instead of traditional convolution. The experimental results show that the algorithm has good generalization while ensuring self-adaptive satisfaction. [ABSTRACT FROM AUTHOR]

The Hyperspectral Imaging System can obtain the spectral response value of each pixel in a fabric sample image while being applied to fabric color measurement. Compared with the common grating Hyperspectral Imaging System, the Hyperspectral Imaging System based on Liquid Crystal Tunable Filter (LCTF) has the advantages of small size, high spatial resolution, and easy integration, which has potential application prospects in textile industry. In this article, the performance of the LCTF Hyperspectral Imaging System for fabric color measurement was evaluated, and an improved model for calibrating the consistency between it and Datacolor 650 was proposed. First, the 210 Pantone cotton color cards were used as samples to study the color repeatability and reproducibility of the LCTF Hyperspectral Imaging System. For evaluation its color measurement ability, the measured reflectance values were converted to the CIELAB color space and compared with those of Datacolor 650. Then, a Radial Basis Function neural network model was proposed to calibrate the color consistency between LCTF Hyperspectral Imaging System and Datacolor 650 which can fully use the reflectance information of the whole band instead of only adjacent wavelengths. Experimental results show that the LCTF Hyperspectral Imaging System can accurately measure the fabric color with good repeatability and reproducibility, and the consistency between the two instruments was significantly improved based on the Radial Basis Function neural network model compared with the R model. [ABSTRACT FROM AUTHOR]

The objective of this foundational study was to develop and evaluate the efficacy of an affordable hyperspectral imaging (HSI) system to identify single and mixed strains of foodborne pathogens in dairy products. This study was designed as a completely randomized design with three replications. Three strains each of Escherichia coli O157:H7 and Listeria monocytogenes were evaluated either as single or mixed strains with the HSI system in growth media and selected dairy products (whole milk, and cottage and cheddar cheeses). Test samples from freshly prepared single or mixed strains of pathogens in growth media or inoculated dairy products were streaked onto selective media (PALCAM and/or Sorbitol MacConkey agar) for isolation. An isolated colony was selected and mixed with 1 ml of HPLC grade water, vortexed for 1 min, and spread over a microscope slide. Images were captured at 2000× magnification on the built HSI system at wavelengths ranging from 400 nm to 1100 nm with 5‐nm band intervals. For each image, three cells were selected as regions of interest (ROIs) to obtain hyperspectral signatures of respective bacteria. Reference pathogen libraries were created using growth media, and then test pathogenic cells were classified by their hyperspectral signatures as either L. monocytogenes or E. coli O157:H7 using k‐nearest neighbor (kNN) and cross‐validation technique in R‐software. With the implementation of kNN (k = 3), overall classification accuracies of 58.97% and 61.53% were obtained for E. coli O157:H7 and L. monocytogenes, respectively. [ABSTRACT FROM AUTHOR]

This study was aimed at exploring the diagnostic value of high-frequency ultrasound imaging based on a fully convolutional neural network (FCN) for peripheral neuropathy in patients with type 2 diabetes (T2D). A total of 70 patients with T2D mellitus were selected and divided into a lesion group (n = 31) and a nonlesion group (n = 39) according to the type of peripheral neuropathy. In addition, 30 healthy people were used as controls. Hypervoxel-based and FCN-based high-frequency ultrasound images were used to examine the three groups of patients to evaluate their diagnostic performance and to compare the changes of peripheral nerves and ultrasound characteristics. The results showed that the Dice coefficient (92.7) and mean intersection over union (mIOU) (82.6) of the proposed algorithm after image segmentation were the largest, and the Hausdorff distance (7.6) and absolute volume difference (AVD) (8.9) were the smallest. The high-frequency ultrasound based on the segmentation algorithm showed higher diagnostic accuracy (94.0% vs. 86.0%), sensitivity (87.1% vs. 67.7%), specificity (97.1% vs. 94.2%), positive predictive value (93.1% vs. 86.7%), and negative predictive value (94.4% vs. 84.0%) (P < 0.05). There were significant differences in the detection values of the three major nerve segments of the upper limbs in the control group, the lesion group, and the nonlesion group (P < 0.05). Compared with the nonlesion group, the patients in the lesion group were more likely to have reduced nerve bundle echo, blurred reticular structure, thickened epineurium, and unclear borders of adjacent tissues (P < 0.05). In summary, the high-frequency ultrasound processed by the algorithm proposed in this study showed a high diagnostic value for peripheral neuropathy in T2D patients, and high-frequency ultrasound can be used to evaluate the morphological changes of peripheral nerves in T2D patients. [ABSTRACT FROM AUTHOR]

Hepatobiliary tumor is one of the common tumors and cancers in medicine, which seriously affects people's lives, so how to accurately diagnose it is a very serious problem. This article mainly studies a diagnostic method of microscopic images of liver and gallbladder tumors. Under this research direction, this article proposes to use convolutional neural network to learn and use hyperspectral images to diagnose it. It is found that the addition of the convolutional neural network can greatly improve the actual map classification and the accuracy of the map, and effectively improve the success rate of the treatment. At the same time, the article designs related experiments to compare its feature extraction performance and classification situation. The experimental results in this article show that the improved diagnostic method based on convolutional neural network has an accuracy rate of 85%–90%, which is as high as 6%–8% compared with the traditional accuracy rate, and thus it effectively improves the clinical problem of hepatobiliary tumor treatment. [ABSTRACT FROM AUTHOR]

Partial fundoplication is an effective treatment for patients with gastro-oesophageal reflux... Amanda Hii 1,2, David S. Liu 1,2, Feruza Kholmurodova 3, Tanya Irvine 3, Tim Bright 1,3,... 1 Oesophagogastric Surgery Unit, Flinders Medical Centre, Adelaide, South Australia, Australi... B Introduction b : Scleroderma patients can exhibit a multitude of symptoms including both dysphagia and gastro-oesophageal reflux disease (GORD). B Results b : At 6-months post-op, bile reflux into the remaining gastric pouch was identified in 14 LOAGB patients (70%), four LSG patients (31%), and one LRYGB patient (5%). Pertinent themes related to AVF care that were strongly illustrated in this study were: i. surgeon-patient relationships were critical to how patients feel about their fistula care, ii. anaesthetic experiences were highly influential in how patients viewed future fistula interventions and iii. feeling cared for by a wider team provided patients with security and safety. CP patients with constant background pain (± exacerbations) had significantly higher frequency of pain attacks, higher opioid use and affective descriptors than CP patients with intermittent pain. B Results b : A total of 528 patients were included in this study, with 283 patients in the pre-guideline group and 245 patients in the post-guideline group. [Extracted from the article]

Convolutional neural networks (CNN) are becoming increasingly popular in modern remote sensing image classification tasks and have exhibited excellent results. For the existing CNN‐based hyperspectral image (HSI) classification methods, most of which extract spatial or spectral features separately by convolution. But nearly all of these methods ignore the fact that the weighted summation of convolution may lead to appear new features in another dimension. To address this issue, a novel multiscale spectral‐spatial cross‐extraction network (MSSCEN) is proposed for HSI classification. Specifically, the proposed MSSCEN introduces spectral‐spatial features cross extraction module (SSCEM), which fed extracted features from previous layer into spatial and spectral extraction branches separately again, so that the changes that occurred in the other domain after each convolution can be fully utilized. In addition, a new independent data augmentation module based on U‐Net is designed to mitigate the problem of limited labelled samples. The paper conducts experiments on three classic hyperspectral datasets and the results demonstrate that the proposed method achieves the best classification accuracy than other state‐of‐the‐art methods. [ABSTRACT FROM AUTHOR]