Search

Concepts like Machine Learning, Data Mining or Artificial Intelligence have become part of our daily life. This is mostly due to the incredible advances made in computation (hardware and software), the increasing capabilities of generating and storing all types of data and, especially, the benefits (societal and economical) that generate the analysis of such data. Simultaneously, Chemometrics has played an important role since the late 1970s, analyzing data within natural science (and especially in Analytical Chemistry). Even with the strong parallelisms between all of the abovementioned terms and being popular with most of us, it is still difficult to clearly define or differentiate the meaning of Machine Learning, Data Mining, Artificial Intelligence, Deep Learning and Chemometrics. This manuscript brings some light to the definitions of Machine Learning, Data Mining, Artificial Intelligence and Big Data Analysis, defines their application ranges and seeks an application space within the field of analytical chemistry (a.k.a. Chemometrics). The manuscript is full of personal, sometimes probably subjective, opinions and statements. Therefore, all opinions here are open for constructive discussion with the only purpose of Learning (like the Machines do nowadays).

First of all, I want to transmit my most humble thanks to all people who believe that I deserve the “2019 Thomas Hirschfeld” award (kindly supported by FOSS) for my work on near-infrared spectroscopy and, especially, applied on hyperspectral images. I must confess that this award caught me by surprise and that I felt a bit overwhelmed when I received it. It is an honour full of respect and responsibility. I have been given the opportunity of writing this article, and I will profit it to express different personal thoughts about general but relevant aspects of near infrared applied to hyperspectral imaging. Also, since I am more a practitioner in chemometrics (or machine learning or data mining, or …) than a developer, I will also include some insights about the beautiful combination of near-infrared hyperspectral image with chemometrics. This article is just a glimpse of constructive criticism with personal thoughts that comes from my little experience in this field. Therefore, and of course, all opinions here are open for constructive discussion with the only purpose of learning (like the machines do nowadays).

Large amount of information in hyperspectral images (HSI) generally makes their analysis (e.g., principal component analysis, PCA) time consuming and often requires a lot of random access memory (RAM) and high computing power. This is particularly problematic for analysis of large images, containing millions of pixels, which can be created by augmenting series of single images (e.g., in time series analysis). This tutorial explores how data reduction can be used to analyze time series hyperspectral images much faster without losing crucial analytical information. Two of the most common data reduction methods have been chosen from the recent research. The first one uses a simple randomization method called randomized sub-sampling PCA (RSPCA). The second implies a more robust randomization method based on local-rank approximations (rPCA). This manuscript exposes the major benefits and drawbacks of both methods with the spirit of being as didactical as possible for a reader. A comprehensive comparison is made considering the amount of information retained by the PCA models at different compression degrees and the performance time. Extrapolation is also made to the case where the effect of time and any other factor are to be studied simultaneously. J.P Cruz-Tirado acknowledges scholarship funding from FAPESP, grant number 2020/09198–1.

Altres ajuts: acords transformatius de la UAB The limitations to assess dental enamel remineralization have been overcome by a methodology resulting from the appropriate combination of synchrotron radiation-based techniques on both, infrared microspectroscopy and micro X-ray diffraction, with the help of specific data mining. Since amelogenin plays a key role in modulating the mineralization of tooth enamel, we propose a controlled ion release for fluorapatite structural ions (Ca 2+, PO 3-, and F -, also including Zn 2+) by using weak acid and weak base ion-exchange resins in the presence of amelogenin to remineralize the surface of etched teeth. This combination provides the necessary ions for enamel remineralization and a guide for crystal growth due to the protein. Remineralized tooth samples were analyzed by applying the indicated methodology. The synchrotron data were treated using principal component analysis and multivariate curve resolution to analyze the mineral layer formed in the presence and absence of amelogenin. The remineralizing treatment created a fluorapatite layer free of carbonate impurities and with a similar orientation to that of the natural enamel thanks to amelogenin contribution.

16 Páginas.-- 7 Figuras.-- 2 Tablas, first_page settings Open AccessArticle VinegarScan: A Computer Tool Based on Ultraviolet Spectroscopy for a Rapid Authentication of Wine Vinegars by Rocío Ríos-Reina 1,2 [ORCID] , Daniel Caballero 3 [ORCID] , Silvana M. Azcarate 4,5, Diego L. García-González 2 [ORCID] , Raquel M. Callejón 1 and José M. Amigo 6,7,* 1 Área de Nutrición y Bromatología, Facultad de Farmacia, Universidad de Sevilla, C/P. García González, 2, ES-41012 Sevilla, Spain 2 Instituto de la Grasa, CSIC, ES-41089 Sevilla, Spain 3 Media Engineering Group, Department of Computer Science and Telematics Engineering, Faculty of Technology, University of Extremadura, Av. De las Letras, S/N, ES-10003 Cáceres, Spain 4 Facultad de Ciencias Exactas y Naturales, Universidad Nacional de La Pampa, Instituto, de Ciencias de la Tierra y Ambientales de La Pampa (INCITAP), Av. Uruguay 151, Santa Rosa 6300, Argentina 5 Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Godoy Cruz 2290, Argentina 6 Ikerbasque, Basque Foundation for Science, María Díaz de Haro, 3, ES-48013 Bilbao, Spain 7 Department of Analytical Chemistry, Faculty of Science and Technology, University of the Basque Country, Barrio Sarriena, S/N, ES-48940 Leioa, Spain * Author to whom correspondence should be addressed. Academic Editor: Elena Benito-Peña Chemosensors 2021, 9(11), 296; https://doi.org/10.3390/chemosensors9110296 Received: 9 September 2021 / Revised: 13 October 2021 / Accepted: 20 October 2021 / Published: 22 October 2021 (This article belongs to the Special Issue Feature Papers on Optical Chemical Sensors and Biosensors) Download PDF Browse Figures Citation Export Abstract Ultraviolet-visible (UV-vis) spectroscopy has shown successful results in the last few years to characterize and classify wine vinegar according to its quality, particularly those with a protected designation of origin (PDO). Due to these promising results, together with the simplicity, price, speed, portability of this technique and its ability to create robust hierarchical classification models, the objective of this work was the development of a computer tool or software, named VinegarScan, which uses the UV-vis spectra to be able to perform quality control and authentication of wine vinegar in a quick and user-friendly way. This software was based on the open-source GUI created in C++ using several data mining algorithms (e.g., decision trees, classification algorithms) on UV-vis spectra. This software achieved satisfactory prediction results with the available analytical UV-vis data. The future idea of utility is to combine the VinegarScan tool with a portable UV-vis device that could be used by control bodies of the wine vinegar industry to achieve a clear differentiation from their competitors to avoid fraud., This research was funded by “Consejería de Economía, Innovación y Ciencia” of “Junta de Andalucía” [P12-AGR-1601].

espanolLa presencia de la cabra montes (Capra pyrenaica) en el Parque Nacional de la Sierra de Guadarrama (Madrid) se remonta a las reintroducciones realizadas a principios de la decada de 1990. Desde entonces, la poblacion ha experimentado un incremento exponencial que ha provocado una elevada presion sobre el medio, especialmente sobre la vegetacion. Desde el ano 2000 se vienen realizando censos periodicos que han permitido conocer la evolucion de la poblacion y su ubicacion dentro del Parque. Con este trabajo se pretende elaborar “mapas de presencia” teniendo en cuenta no solo la situacion actual sino tambien la presencia historica y la continuidad de la especie dentro del parque. A partir de los censos realizados y considerando 5 variables diferentes (numero de individuos presentes, estacionalidad, numero de anos con presencia, incremento poblacional y presencia estos ultimos anos) se han determinado cuales han sido las zonas que han tenido una presencia mayor y mas continuada de la especie desde su reintroduccion. Se han establecido 5 zonas dentro del Parque Nacional, siendo las zonas de La Najarra y Cuerda Larga las que muestran una mayor presencia y continuidad, las zonas aledanas de Pedriza Anterior y Posterior muestran una presencia relativamente alta, la cuerda de Montes Carpetanos tiene una presencia mas reciente y es nula en zonas perifericas del Parque (Siete Picos y cuerda de Navarredonda). EnglishThe presence of the Iberian Ibex (Capra pyrenaica) in the Parque Nacional Sierra de Guadarrama (Madrid) dates back to the reintroductions made in the early 1990s. Since then, the population has experienced an exponential increase causing high pressure on the environment, especially on the vegetation. Periodic censuses have been carried out from 2000 which have allowed us to track the evolution of the population and its location within the Park. The aim of this work is to draw up ‘presence maps’ taking into account not only the current situation but also the historical presence and the continuity of the species within the National Park. Based on censuses carried out since 2000 and considering 5 different variables -number of individuals present, seasonality, number of years with presence, population increase and presence in last years-, the areas that have had the greatest and most continuous presence of the species has been determined since its reintroduction. There have been established five zones within the National Park, showing the area of La Najarra and Cuerda Larga a greater presence and continuity, the surrounding areas of Pedriza Anterior and Pedriza Posterior a moderate presence, the Montes Carpetanos rope shows a lower presence and is null in peripheral areas of the Park (Seven Picos and Navarredonda rope).

In recent years, three Spanish wine vinegars have obtained the indication of Protected Denomination of Origin (PDOs) due to their unique characteristics and traditional method of production: “Vinagre de Jerez”, “Vinagre de Condado de Huelva” and “Vinagre de Montilla-Moriles”. These vinegars are expensive due to their high quality, the long aging time and the high cost of production, reason why the adulteration and unfair competition in the vinegar industry are frequent practices. To avoid these frauds, several analytical techniques have been already studied for the characterization and authentication of these high quality vinegars. Nevertheless, ultraviolet–visible (UV–vis) spectroscopy, especially attractive for its simplicity and low cost, has not been previously used to assess PDO or other qualities as type of production or aging, in wine vinegars. For this reason, the potential of UV–vis spectroscopy was investigated for the first time as a rapid and inexpensive methodology for developing classification models to discriminate wine vinegars according to the production method, the PDO and the aging category. Spectra from 70 wine vinegars -including different categories within the 3 PDOs and also vinegars without PDO as known as rapid vinegars-have been analyzed and compared in the selected region of 280–600 nm. Principal components analysis (PCA) was used as exploratory method, while soft independent modelling-class (SIMCA) and partial least squares-discriminant analysis (PLS-DA) were employed for the development of a hierarchical classification model. Differences between categories and PDOs, as well as between PDO and Non-PDO wine vinegars, were observed according to the spectral regions around 300 nm and the visible regions around 500 nm. Furthermore, bootstrap resampling method was employed to generate distributions of classification results and to obtain confidence intervals in the classification. The hierarchical classification results open up the possibility of developing a tool that provides an easy and fast differentiation for the authentication of wine vinegars from different categories and denomination of origins. Fil: Ríos Reina, Rocío. Universidad de Sevilla; España Fil: Azcarate, Silvana Mariela. Consejo Nacional de Investigaciones Científicas y Técnicas. Instituto de Ciencias de la Tierra y Ambientales de La Pampa. Universidad Nacional de La Pampa. Facultad de Ciencias Exactas y Naturales. Instituto de Ciencias de la Tierra y Ambientales de La Pampa; Argentina Fil: Camiña, José Manuel. Consejo Nacional de Investigaciones Científicas y Técnicas. Instituto de Ciencias de la Tierra y Ambientales de La Pampa. Universidad Nacional de La Pampa. Facultad de Ciencias Exactas y Naturales. Instituto de Ciencias de la Tierra y Ambientales de La Pampa; Argentina Fil: Callejón, Raquel M.. Universidad de Sevilla; España Fil: Amigo, José Manuel. Universidad de Copenhagen; Dinamarca

This work addresses the industrial process of compounding automotive gasoline by a controlled blending of feedstock hydrocarbon streams deriving from different stages of the cracking and refining process of naphtha. In this process, gasoline quality and profitability are directly affected by both the composition of each stream and the ratio of the streams to each other in the gasoline. Indeed, factors such as the process conditions, feedstock supply, production costs and origin of the naphtha frequently vary in both composition and ratio. As a result, the gasoline blending recipes must change constantly over time to keep up with the quality and economic characteristics of the product, with octane rating as a critical operational parameter in these processes. Near Infrared (NIR) spectroscopy is able to detect changes in composition and it can be considered during the design of a gasoline. Moreover, the use of NIR spectroscopy makes it possible to reduce significantly the time needed for analyzing the parameters of interest in the process, compared to the analytical methods that are commonly used. The main purpose of this work was to develop an off-line simulator of octane number in automotive gasoline based on NIR spectroscopy and chemometrics, aiming to improve the compounding process. The simulator was developed on the basis of the mathematical reproduction of gasoline spectra by a virtual controlled mixture of the near infrared (NIR) spectra of the hydrocarbon streams at appropriate ratio. Spectra were recorded with two instruments, a benchtop FT-NIR and a handheld instrument. Partial Least Squares (PLS) regression models were developed and validated to estimate the octane number from the simulated NIR spectra for potential gasolines. The comparison among the octane number predictions from the simulated spectra, the real NIR spectra and, also, the reference values, at different times over 2 years of production, representing a critical step in the development of the simulator, showed that the simulated spectra were able to predict values reliably. The developed octane simulator has an excellent potential for improving efficiency in the compounding process of automotive gasoline.

The complex physical transformations of polymers upon external thermodynamic changes are related to the molecular length of the polymer and its associated multifaceted energetic balance. The understanding of subtle transitions or multistep phase transformation requires real-time phenomenological studies using a multi-technique approach that covers several length-scales and chemical states. A combination of X-ray scattering techniques with Raman spectroscopy and Differential Scanning Calorimetry was conducted to correlate the structural changes from the conformational chain to the polymer crystal and mesoscale organization. Current research applications and the experimental combination of Raman spectroscopy with simultaneous SAXS/WAXS measurements coupled to a DSC is discussed. In particular, we show that in order to obtain the maximum benefit from simultaneously obtained high-quality data sets from different techniques, one should look beyond traditional analysis techniques and instead apply multivariate analysis. Data mining strategies can be applied to develop methods to control polymer processing in an industrial context. Crystallization studies of a PVDF blend with a fluoroelastomer, known to feature complex phase transitions, were used to validate the combined approach and further analyzed by MVA Région Grand’EST, Arkema; Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO).

[EN] Pasta is mostly composed by wheat flour and water. Nevertheless, flour can be partially replaced by fibers to provide extra nutrients in the diet. However, fiber can affect the technological quality of pasta if not properly distributed. Usually, determinations of parameters in pasta are destructive and time-consuming. The use of Near Infrared-Hyperspectral Imaging (NIR-HSI), together with machine learning methods, is valuable to improve the efficiency in the assessment of pasta quality. This work aimed to investigate the ability of NIR-HSI and augmented Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS) for the evaluation, resolution and quantification of fiber distribution in enriched pasta. Results showed R2V between 0.28 and 0.89, %LOF < 6%, variance explained over 99%, and similarity between pure and recovered spectra over 96% and 98% in models using pure flour and control as initial estimates, respectively, demonstrating the applicability of NIR-HSI and MCR-ALS in the identification of fiber in pasta., This work was supported by the Coordenaçao de Aperfeicoamento de Pessoal de Nivel Superior -Brasil (CAPES) [Finance Code 001]; Sao Paulo Research Foundation (FAPESP) [grant numbers 2008/57808-1, 2014/50951-4, 2015/24351-2, 2017/17628-3, 2019/06842-0]; and by GVA-IVIA and FEDER funds through project IVIA-51918. The authors would like to thank Nutrassim Food Ingredients company for the donation of the fiber samples, the support provided by Mrs. Cristiane Vidal during NIR-HSI system operation and data processing and Dr. Celio Pasquini for promptly receiving us in the laboratory that he coordinates (Grupo de instrumentaçao e automaçao em quimica analitica, Instituto de quimica, Universidade Estadual de Campinas, Campinas-SP, Brazil) to data acquisition.

Microplastics are defined as microscopic plastic particles in the range from few micrometers and up to 5 mm. These small particles are classified as primary microplastics when they are manufactured in this size range, whereas secondary microplastics arise from the fragmentation of larger objects. Microplastics are widespread emerging pollutants, and investigations are underway to determine potential harmfulness to biota and human health. However, progress is hindered by the lack of suitable analytical methods for rapid, routine, and unbiased measurements. This work aims to develop an automated analytical method for the characterization of small microplastics (

This paper explores how the staling of white bread affects the behavior of the whole crumb surface and how that mechanism is interrupted/changed by the addition of maltogenic α-amylases. This is done using near infrared hyperspectral imaging, machine learning methodologies and the knowledge acquired in the previous two manuscripts. Methods like principal component analysis and multivariate curve resolution demonstrate how the constituents of the bread being stored (for 21 days) evolve differently depending on the presence/absence of maltogenic α-amylases and also which parts of the crumb are primarily exposed to changes. The spatial distribution of the hardness is calculated in the entire surface of the slice area during staling by using partial least square regression. This manuscript comprehends one of the largest studies made on white bread staling and proposes a complete methodology using near infrared hyperspectral imaging and machine learning.

Fourier Transform Mid Infrared with Attenuated Total Reflection Imaging (FTIR-ATR imaging) and Multivariate Curve Resolution with Alternating Least Squares (MCR-ALS) were used in a multiblock fashion to study the presence, distribution and penetration depth of very low concentrations of florfenicol (FF) in a complex matrix like feed pellets for salmonids. Images from the surface, at 150 µm deep and 200 µm deep from the surface were analyzed to certify the penetration power of FF added by surface coating methodology. Besides, the unique homogeneity index was calculated in order to evaluate the distributional homogeneity of each component. The results demonstrated the reliability of MCR-ALS in studying the distributional homogeneity of FF. It was demonstrated that FF remains mostly on the surface of the pellets with almost no penetration. The rest of the components of the pellets (oil, protein and carbohydrates) were also analyzed. These three nutrients are distributed on the three layers analyzed with a relatively homogeneous location, being carbohydrates (%H = 51 ± 3) the component with the best homogeneous distribution, unlike protein (%H = 45 ± 5), and oil (%H = 40 ± 7). This is the first publication where the penetration of an antibiotic, added with surface-coating to feed pellets, was analyzed with FTIR-ATR imaging and multivariate analysis, showing the contribution these analytical tools can make to the medicated feed industry.

This paper demonstrates that the use of the homogeneity curve approach introduced in the first part of this work for assessing the unique homogeneity percentage (%H) can also be applied in single channel non-binary (SCNB) images (a.k.a grayscale images) analysis. In order to address a more general homogeneity index approach for SCNB images, more realistic assumptions such as the presence of divisible objects or the mixture of different objects in the same pixel, are considered. The proposed parameter, %H, is an absolute homogeneity percentage that uses only the self-contained information of one image, with no need for additional modelling steps, as described in most of the works addressing the homogeneity assessment. To demonstrate it, two cases have been evaluated. The first case validates the %H by assessing the homogeneity in images with different textures (distribution of objects). In the second case, the reliability of assessing the evolution of homogeneity and providing an end-point detection in blending process trials done at small scale has been evaluated. The results obtained in the current work show the high potential when using the %H for industrial applications where the distributional homogeneity represents a critical process parameter to ensure the quality of the final product.

Computer vision algorithms on Magnetic Resonance Imaging have been introduced as an alternative to destructive methods for determining the quality traits of meat products, since Magnetic Resonance Imaging is a non-destructive, non-ionizing, and innocuous method. The use of fractal properties to analyze Magnetic Resonance Imaging could be another possibility for this purpose. In this paper, a new fractal properties algorithm was developed, to obtain features from Magnetic Resonance Imaging based on fractal characteristics. This algorithm is called One Point Fractal Texture Algorithm. The fractal properties of this new algorithm were compared with those of other fractal (Classical Fractal Algorithm and Fractal Texture Algorithm) and the classical texture ones (Grey level co-occurrence matrix, grey level run length matrix and neighbouring gray level dependence matrix). The results obtained by means of these computer vision algorithms were correlated to the results obtained by means of physico-chemical and sensory analysis. Classical Fractal Algorithm reached low association for the quality parameters of loins. The remaining algorithms achieved correlation coefficients higher than 0.5, noting One Point Fractal Texture Algorithm that reached the highest correlation coefficients in all cases except for the L* coordinate color. These high correlation coefficients confirm the new algorithm as an alternative to the other computer vision approaches in order to compute the physico-chemical and sensory parameters of meat products in a non-destructive and efficient way.

Industry requires non-destructive real-time methods for quality control of meat in order to improve production efficiency and meet consumer expectations. Near Infrared Hyperspectral Images were used for tenderness evaluation of Nellore beef and the construction of tenderness distribution maps. To investigate whether the selection of the region of interest (ROI) in the image at the exact location where the shear force core was collected improves tenderness prediction and classification models, 50 samples from Longissimus muscle were imaged (1000–2500 nm) and shear force were measured (Warner-Bratzler). The data were analyzed by chemometric techniques (Partial Least Squares together with discriminant analysis – PLS-DA). Classification models using local ROI presented better performance than the ROI models of the whole sample (external validation sensitivity for the tough class = 33% and 70%, respectively), but none could be considered as successful model. However, the more general model had better performance in the tenderness distribution maps, with 72% of predicted images correctly classified.

3 Tablas.-- 3 Figuras.-- 2 Figuras suplementarias, High-quality wine vinegars protected by the indication “Protected Designation of Origin” (PDO) need efficient tools to protect their brands and prevent adulteration and unfair competition. In this sense, Near-Infrared spectroscopy (NIRs) combined with chemometrics has demonstrated its usefulness in food authentication. This work assessed NIRs and Chemometrics as a rapid and non-destructive methodology for this purpose. In this study, 83 high-quality wine vinegars of the Spanish PDOs “Vinagre de Jerez” “Vinagre de Condado de Huelva” and “Vinagre de Montilla-Moriles” of different categories, and 11 wine vinegars without PDO, were analyzed in the range 12000-4000 cm. Principal component analysis (PCA) was performed to explore the spectra and Partial Least Squares-Discriminant Analysis (PLS-DA) was used to build classification models. The high ability of prediction obtained (>90% correct classification) demonstrated the usefulness of this methodology for authentication of PDO wine vinegars and their categories., The authors would like to thank the Spanish Regulatory Councils of the wine vinegars PDOs for their invaluable help with the acquisition of the samples for the study. Authors also thank University of Copenhagen, specifically to Professor Franciscus Winfried J van der Berg and the department of food science for providing the equipment and their knowledge in the methodology. This work was supported by “Consejeria de Economía, Innovación y Ciencia” of “Junta de Andalucia” [P12-AGR-1601], and “Ministerio de Educación Cultura y Deporte” (MECD, Spanish Government) for the FPU pre-doctoral fellowship [grant number FPU2014/01247].

This study investigates the use of fractal algorithms to analyse MRI of meat products, specifically loin, in order to determine sensory parameters of loin. For that, the capability of different fractal algorithms was evaluated (Classical Fractal Algorithm, CFA; Fractal Texture Algorithm, FTA and One Point Fractal Texture Algorithm, OPFTA). Moreover, the influence of the acquisition sequence of MRI (Gradient echo, GE; Spin Echo, SE and Turbo 3D, T3D) and the predictive technique of data mining (Isotonic regression, IR and Multiple Linear regression, MLR) on the accuracy of the prediction was analysed. Results on this study firstly demonstrate the capability of fractal algorithms to analyse MRI from meat product. Different combinations of the analysed techniques can be applied for predicting most sensory attributes of loins adequately (R > 0.5). However, the combination of SE, OPFTA and MLR offered the most appropriate results. Thus, it could be proposed as an alternative to the traditional food technology methods.

[EN] Product inspection is essential to ensure good quality and to avoid fraud. New nectarine cultivars with similar external appearance but different physicochemical properties may be mixed in the market, causing confusion and rejection among consumers, and consequently affecting sales and prices. Hyperspectral reflectance imaging in the range of 450¿1040 nm was studied as a non-destructive method to differentiate two cultivars of nectarines with a very similar appearance but different taste. Partial least squares discriminant analysis (PLS-DA) was used to develop a prediction model to distinguish intact fruits of the cultivars using pixel-wise and mean spectrum approaches, and then the model was projected onto the complete surface of fruits allowing visual inspection. The results indicated that mean spectrum of the fruit was the most accurate method, a correct discrimination rate of 94% being achieved. Wavelength selection reduced the dimensionality of the hyperspectral images using the regression coefficients of the PLS-DA model. An accuracy of 96% was obtained by using 14 optimal wavelengths, whereas colour imaging and a trained inspection panel achieved a rate of correct classification of only 57% of the fruits., This work was partially funded by INIA and FEDER funds through project RTA2015-00078-00-00. Sandra Munera thanks INIA for the FPI-INIA grant num. 43 (CPR2014-0082), partially supported by European Union FSE funds. The authors wish to thank Fruits de Ponent (Lleida) for providing the fruit.

HYPER-Tools is a new graphical user-friendly interface (GUI) especially designed for the analysis of hyperspectral images. This easy-to-use interface works under Matlab environment and integrates fundamental types of spectral and spatial pre-processing methods as well as the main chemometric tools (exploratory data analysis, clustering, regression, and classification) for the comprehensive analysis of hyperspectral images. The main features of HYPER-Tools are the powerful visualization tools implemented, the availability of working with several images simultaneously and the easiness of the interaction of the user with the interface. Together with the GUI, several tutorials and videos are provided in the official website ( https://www.hypertools.org/ ) showing the working procedure of HYPER-Tools step by step in different situations.

Analysis of gunshot residues (GSR) is a vital point of interest in forensic chemistry. GSR analysis is very well established, but there is still a lack of methodology to identify GSR obtained from lead-free or ‘non-toxic ammunition’ (NTA). To identify GSR from NTA, we have proposed the incorporation of luminescent markers to provide an optical and chemical signature to GSR. Furthermore, the luminescent markers can open up prospectives for ammunition encoding by adjusting their composition. Herein, we report the distinction among GSR particles containing different MOF-based markers by analyzing their organic fractions using near-infrared hyperspectral imaging (NIR-HSI) together with multivariate analysis (MCR-ALS). The method has been proven to be effective, non-destructive and fast in distinguishing marked GSR. It is possible to detect and correctly identify 72.2% of the samples, and all cases of misclassification are related to slight amount of material collected (8.3%) or confusion caused by one of the markers (19.4%). The study is developed in a forensic context of GSR analysis; however, the methodology described here is excellent and can be extrapolated to other applications that require distinguishing MOFs with different ligands.

Multispectral images of clothing targets shot at seven different distances (from 10 to 220 cm) were recorded at 18 specific wavelengths in the 400&-1000 nm range to visualize the gunshot residue (GSR) pattern. Principal component analysis (PCA) showed that the use of violet-blue wavelengths (430, 450 and 470 nm) provided the largest contrast between the GSR particles and the white cotton fabric. Then, the correlation between the amount of GSR particles on clothing targets and the shooting distance was studied. By selecting the blue frame of multispectral images (i.e. the blue frame in the red-green-blue (RGB) system which falls at 470 nm), the amount of pixels containing GSR particles was accounted based on the intensity of pixels in that frame. Results demonstrated that the number of pixels containing GSR exponentially decreases with the shooting distance from 30 to 220 cm following a particular exponential equation. However, the targets shot at the shortest distance (10 cm) did not satisfy the above equation, probably due to the noticeable differences of the GSR-pattern of these targets (e.g. high presence of soot). Then, the equation was applied to validation samples to estimate the shooting distances, obtaining results with an error below 10%.

2 Tablas.-- 6 Figuras, The use of data fusion methodologies has increased at the same rhythm as the capability of modern analytical laboratories of measuring sample from multiple sources. Almost all data fusion strategies can be grouped into three levels, they fuse the data differently with the sole aim of obtaining a better response (qualitative or quantitative) than that obtained by the instruments individually. One of the major key points for the data fusion methodologies to succeed is the understanding of the data structure obtained from a particular instrument. This point is not exhaustively commented in the literature focused on data fusion, sometimes paying too much attention to the algorithms instead. This manuscript explains data fusion from the structure of the different data obtained by different analytical platforms. Special attention will be given to the nature of the data and the relationships between the samples and the variables, as well as within the variables., The authors are grateful to Univ. Nacional de La Pampa, and CONICET (Consejo Nacional de Investigaciones Científicas y Técnicas, (Project PIP-2015 Nº 0111) and ANPCyT (Agencia Nacional de Promoción Científica y Tecnológica, (Projects PICT 2017-0340 and PICT-2018-04496) for financial support.

Texture features analysis is one of the most important approaches for the assessment of homogeneity on images. However, all of them are either relative to the comparison with a standardized set of images, or further multivariate models are strongly required to predict or classify the images according to their features. In this first work, we propose an alternative and novel methodology to calculate a percentage of homogeneity by only using the self-information contained on the image. This methodology is based on the macropixel analysis theory and the generation of what is called the “homogeneity curve”. The homogeneity curve is deeply explored and the knowledge to what it could be considered the most homogeneous and inhomogeneous distribution for every case is spanned. This first work postulates the theory and demonstrates its usefulness with several examples applied to binary images. This will provide a theoretical framework to fully understand the homogeneity curve, postulating a mathematical model to parametrize homogeneity and its plausible deviations.

Identification of semen stains is a critical step in evidence analysis in cases of sexual assaults (rape, child molestation, sexual harassments, etc). Presumptive and confirmatory methodologies that are nondestructive and can be used at crime scenes are extremely needed to help forensic experts to collect useful evidences without compromising the integrity of the samples. In this work, we put forward the high potential of a fast, nondestructive and noninvasive methodology like Near Infrared Hyperspectral Imaging (HSI-NIR), together with adequate multivariate methodologies for exploratory data analysis and classification. Due to the high absorption of the substrate, an exhaustive data analysis is performed in order to test the most suitable tool for the detection and discrimination of semen stains in diverse fabrics. Samples of human semen, lubricants, breast milk and animal semen were placed on fabrics of different colors and compositions to create a stain and hyperspectral images were acquired. As identification tools, Principal Component Analysis and MCR-ALS were used. Moreover, Classification techniques such as PLS-DA, sPLS-DA and SVM-DA were used in order to differentiate human semen from other substances. Although the methodology developed depend on the substrate used, it has been proven that HSI-NIR can be used as presumptive and confirmatory method of analysis in the search for semen stains.

This work firstly investigates the use of MRI, fractal algorithms and data mining techniques to determine pork quality parameters non-destructively. The main objective was to evaluate the capability of fractal algorithms (Classical Fractal algorithm, CFA; Fractal Texture Algorithm, FTA and One Point Fractal Texture Algorithm, OPFTA) to analyse MRI in order to predict quality parameters of loin. In addition, the effect of the sequence acquisition of MRI (Gradient echo, GE; Spin echo, SE and Turbo 3D, T3D) and the predictive technique of data mining (Isotonic regression, IR and Multiple linear regression, MLR) were analysed. Both fractal algorithm, FTA and OPFTA are appropriate to analyse MRI of loins. The sequence acquisition, the fractal algorithm and the data mining technique seems to influence on the prediction results. For most physico-chemical parameters, prediction equations with moderate to excellent correlation coefficients were achieved by using the following combinations of acquisition sequences of MRI, fractal algorithms and data mining techniques: SE-FTA-MLR, SE-OPFTA-IR, GE-OPFTA-MLR, SE-OPFTA-MLR, with the last one offering the best prediction results. Thus, SE-OPFTA-MLR could be proposed as an alternative technique to determine physico-chemical traits of fresh and dry-cured loins in a non-destructive way with high accuracy.

This study explores the effect of different settings on beer fermentation process applying an interval-based version of ASCA on FT-IR data. Three main factors (yeast type, temperature, fermentation time) are included in the experimental design, being high sources of variation in brewing and strictly interdependent; thus, difficult to be studied through a univariate approach. The three-factor full factorial design leads to a spectral multi-set data, with a total of 12 independent fermentations, which is explored combining ASCA and an interval adaptation of ASCA (interval-ASCA or i-ASCA). The ASCA models calculated on two separate regions (2900–2250 cm −1 and 1500–980 cm −1 ) shows differences for average time levels and the interaction between yeast types and time linked to carbon dioxide, maltose consumption and ethanol production, respectively. To better investigate the punctual influence of the studied factors on the so-called IR fingerprint region, permutation testing of ASCA in variable intervals is investigated. The analysis highlights the significant effect not only of the fermentation in all intervals considered, but also the role of other factors, such as time × yeast, yeast and temperature, in smaller variable regions. The proposed approach demonstrates how interval-ASCA on FT-IR data, isolating the variation in the data according to the experimental design used, allows a rapid and accurate test for parameter control in beer manufacturing.

Remote identification of illegal plantations of Cannabis sativa Linnaeus is an important task for the Brazilian Federal Police. The current analytical methodology is expensive and strongly dependent on the expertise of the forensic investigator. A faster and cheaper methodology based on automatic methods can be useful for the detection and identification of Cannabis sativa L. in a reliable and objective manner. In this work, the high potential of Near Infrared Hyperspectral Imaging (HSI-NIR) combined with machine learning is demonstrated for supervised detection and classification of Cannabis sativa L. This plant, together with other plants commonly found in the surroundings of illegal plantations and soil, were directly collected from an illegal plantation. Due to the high correlation of the NIR spectra, sparse Principal Component Analysis (sPCA) was implemented to select the most important wavelengths for identifying Cannabis sativa L. One class Soft Independent Class Analogy model (SIMCA) was built, considering just the spectral variables selected by sPCA. Sensitivity and specificity values of 89.45% and 97.60% were, respectively, obtained for an external validation set subjected to the s-SIMCA. The results proved the reliability of a methodology based on NIR hyperspectral cameras to detect and identify Cannabis sativa L., with only four spectral bands, showing the potential of this methodology to be implemented in low-cost airborne devices.

Precision agriculture is starting to play a key role in the new generation of modern agricultural revolution. Farmers are increasingly aware of the importance of maintaining direct control of several fundamental aspects such as the state of health of crops, the amount of water or fertilizer, and possible infections that can develop in the field. Hyperspectral imaging (HSI) and multispectral imaging (MSI) have been applied for these matters for some decades. Nevertheless, there are still several technological and practical barriers to face. This chapter provides an overview of some of the relevant scientific literature related to the application of HSI and MSI on crop fields. Some of the applications are related to the detection of contaminants and heavy metals, the control and assessment of water sources, and the detection of the health status of the fields. The main advantages and constraints and the methods to acquire the images (by using satellites or aerial vehicles) are also shown and discussed.

Hyperspectral and multispectral imaging (HSI and MSI, respectively) are well-known imaging techniques that have their origin in remote sensing. Nevertheless, the versatility of having a camera with the integration of a spectrometer for each pixel has profited many other branches of science and industry. Nowadays they are used in precision agriculture, pharmaceutical production, artworks, biochemistry, etc. There are several fields in which their implementation is still growing and being adapted to the requirements of the field. This chapter will review some of them, such as their implementation in waste sorting, forensic sciences, archaeology, or entomology. Of course, there are many other fields of expertise taking benefits from HSI and MSI. These are just some examples of unusual applications of HSI and MSI where we believe that there is still a vast potential for growth.

Audio fingerprinting involves extraction of quantitative frequency descriptors that can be used for indexing, search and retrieval of audio signals in sound recognition software. We propose a similar approach with medical ultrasonographic Doppler audio signals. Power Doppler periodograms were generated from 84 ultrasonographic Doppler signals from the common carotid arteries in 22 dogs. Frequency features were extracted from each periodogram and included in a principal component analysis (PCA). From this 10 audio samples were pairwise classified as being either similar or dissimilar. These pairings were compared to a similar classification based on standard quantitative parameters used in medical ultrasound and to classification performed by a panel of listeners. The ranking of sound files according to degree of similarity differed between the frequency and conventional classification methods. The panel of listeners had an 88% agreement with the classification based on quantitative frequency features. These findings were significantly different from the score expected by chance (p

[EN] The internal quality of nectarines (Prunus persica L. Batsch var. nucipersica) cv. 'Big Top' (yellow flesh) and 'Magique' (white flesh) has been inspected using hyperspectral transmittance imaging. Hyperspectral images of intact fruits were acquired in the spectral range of 630-900 nm using transmittance mode during their ripening under controlled conditions. The detection of split pit disorder and classification according to an established firmness threshold were performed using PLS-DA. The prediction of the Internal Quality Index (IQI) related to ripeness was performed using PLS-R. The most important variables were selected using interval-PLS. As a result, an accuracy of 94.7% was obtained in the detection of fruits with split pit of the 'Big Top' cultivar. Accuracies of 95.7% and 94.6% were achieved in the classification of the 'Big Top' and 'Magique' cultivars, respectively, according to the firmness threshold. The internal quality was predicted through the IQI with R-2 values of 0.88 and 0.86 for the two cultivars. The results obtained indicate the great potential of hyperspectral transmittance imaging for the assessment of the internal quality of intact nectarines., This work was partially funded by INIA and FEDER funds through project RTA2015-00078-00-00. Sandra Munera thanks INIA for the FPI-INIA grant num. 43 (CPR2014-0082), partially supported by European Union FSE funds.

This second paper provides chemical insight of the different phenomena occurring in bread during storage with and without anti-staling enzymes by using near infrared spectroscopy and chemometrics. The target, thus, is three-fold: (1) To monitor the staling process in the top, middle and bottom parts of the white bread loaf by near infrared spectroscopy and to extract chemical information of the different chemical mechanisms occurring in the staling process; (2) to assess the correlation between the near infrared spectroscopy and spatial texture profile analysis in terms of hardness, and (3) to demonstrate the anti-staling effect of the enzymes by showing a collapse of the correlation between near infrared (NIR) spectra and hardness as measured by texture analysis. It is found that NIR spectroscopy in combination with chemometrics (Partial Least Squares Regression) can predict the hardness development of the control bread.

A new methodology based on Near Infrared-hyperspectral imaging and Principal Components Analysis (PCA) was developed and accurately validated to model the degradations kinetics and to estimate the multivariate accelerated shelf life (MASLT) of chia seeds (Salvia hispanica). Chia seeds were stored 180 days at 25, 35 and 45 °C, observing fatty acid degradation and an increase in the acidity. PC1 scores and kinetic charts were built fitting the time-related PC1 scores versus time by a fused kinetic model (R2 > 0.85). The spectra of chia seeds where acidity increased at 75% from initial value were used to calculate the cut-off value (−0.9853). The shelf life estimations were 1300, 798 and 90 days for chia seeds stored at 25, 35 and 45 °C, respectively. For the first time, a reliable methodology is proposed to validate that all samples were correctly predicted using PC1 scores.

Solid-state transitions at the surface of pharmaceutical solid dosage forms (SDF) were monitored using multi-series hyperspectral imaging (HSI) along with Multivariate Curve Resolution – Alternating Least Squares (MCR-ALS) and Parallel Factor Analysis (PARAFAC and PARAFAC2). First, the solid-state transformation due to the dehydration of the monohydrate forms of piroxicam and lactose to their respective anhydrate counterparts in tablets were monitored using temperature series NIR-HSI. PARAFAC and MCR-ALS solutions were hampered by the lack of strict trilinearity of the pixels among the unfolded series NIR-images (three-way array) and due to rotational ambiguity (augmented matrix), respectively, while PARAFAC2 resolved satisfactorily the profile of the corresponding compounds in the pixels in the series NIR-images. Next, the amorphous-to-crystalline transitions were monitored in solid dispersion of indomethacin with polyvinylpyrrolidone using time series MIR-HSI. MCR-ALS properly resolved the known solid-state forms of the drug in the pixels of the series MIR-images, while PARAFAC and PARAFAC2 failed to properly resolve all the drug forms in the series MIR-images due to i) strict trilinearity leak in the three-way array and ii) the mandatory constant cross-product AkTAk over the k series MIR-images (A is the loadings of the shift mode), respectively. The highlighting of the advantages and limitation of the corresponding curve resolution methods stressed their potential applicability when handling multi-series HSI to study solid-state transitions in pharmaceutical SDFs.

Physical stability and textural properties of eighteen low fat stirred yoghurts, differing in type of added dairy protein ingredients and in exopolysaccharide (EPS)-producing cultures, were studied by instrumental and sensorial measurements. Image analysis was performed to quantify the amount of grains and the amount of EPS. Enhanced textural properties and physical stability was observed in systems containing whey protein concentrate or particulated whey proteins as well as EPS-producing cultures. Irrespective of the culture used, casein and fractionated whey protein did not contribute to the textural properties. Sensory analysis confirmed the results obtained by instrumental measurements. Oral gel firmness was highly correlated to G′, viscosity at low shear rate, and instrumental measured firmness (R ≈ 0.82). Mouth thickness was well described by viscosity at high shear rate (R = 0.81), as well as by the hysteresis loop area and ropiness (R ≈ 0.81). Measured syneresis showed a significant relationship with visually evaluated syneresis (R = 0.76), and the amount of grains also correlated with visual evaluated graininess (R ≈ 0.81). The effect of the cultures on the sensory properties was revealed by multilevel simultaneous component analysis (MLSCA-P).