La mission JUICE de l’ESA sera la troisième mission d’exploration entièrement dédiée au système de Jupiter, et la première à se concentrer sur les lunes Galiléennes glacées susceptibles d’abriter des océans d’eau liquide. Prévue pour un lancement en 2022 et une insertion en orbite jovienne fin 2029, la sonde emportera parmi ses 11 instruments le spectro-imageur MAJIS. Les données d’un tel instrument comprennent une image à haute résolution spatiale de la zone étudiée et un spectre pour chacun des pixels de cette image. Ce spectre, dans la gamme allant de 0.5 à 5.5 µm, permet d’obtenir des informations physico-chimiques sur le contenu du pixel concerné. Le laboratoire où j’ai effectué mon travail de thèse, l’IAS, s’est vu confier la responsabilité de la réalisation de MAJIS. Dans ce contexte, l’objectif de mon travail était de contribuer à la définition et à l’implémentation de l’étalonnage de l’instrument : j’ai pour cela dû comprendre d’abord ses objectifs scientifiques et les exigences instrumentales qui en découlent, et maîtriser les caractéristiques des sous-systèmes qui composent MAJIS. J’ai tout d’abord traité les données de l’imageur intégral de champ de SPHERE, un instrument du VLT, qui avait observé la lune Galiléenne volcanique Io en 2014. Bien que ce satellite soit un objectif mineur de la mission JUICE, j’ai dû me confronter au fonctionnement de l’instrument pour en réduire les données et le traitement des spectres a requis le développement d’un modèle photométrique d’observation de la surface que j’ai pu confronter à la réalité et à d’autres études. L’identification de nombreux biais systématiques dans ces données et la quantification de ses limites de détection spatiales et spectrales m’ont permis de souligner l’aspect critique de la phase d’étalonnage de MAJIS pour que ses données soient exploitables. Avant cette étape toutefois, la connaissance des sous-systèmes qui vont constituer l’instrument est elle aussi nécessaire car certains de leurs paramètres conditionneront le déroulement de cet étalonnage et ils ne pourront pas tous être mesurés à cette occasion. J’ai donc caractérisé, à l’aide des bancs optiques dédiés à l’IAS, le plan focal de l’instrument et surtout son détecteur CMOS infrarouge de type HgCdTe. J’ai pu mesurer ses caractéristiques les plus courantes, comme son courant d’obscurité, sa profondeur de puits, son efficacité quantique, son éventuelle persistance, son bruit de lecture et la linéarité de sa réponse. Dans le cas d’une mission vers Jupiter, un autre aspect des performances du détecteur doit être étudié en détail : sa résistance aux radiations, particulièrement intenses dans la magnétosphère jovienne. J’ai pu effectuer une série de tests sur des détecteurs témoins avec des sources d’électrons, de protons et de photons de hautes énergies, qui m’ont permis de montrer la très bonne résistance du plan focal aux dégâts permanents. Ces données ont aussi permis de caractériser expérimentalement le signal transient induit par un bombardement aux électrons, ce qui m’a permis de valider l’approche de filtrage de ce signal qui sera implémentée en vol. C’est enfin grâce aux résultats de ces trois approches et au développement d’un modèle photométrique complet de l’instrument et de son dispositif d’étalonnage, que j’ai pu discuter l’architecture de ce dernier et proposer des séquences de mesure pour la campagne d’étalonnage. J’ai donc travaillé avec les ingénieurs du laboratoire et des industriels pour réaliser ce dispositif d’étalonnage, sélectionner les sources de lumière qui permettront la mesure de la réponse spatiale, spectrale et radiométrique de l’instrument nécessaires à l’interprétation de ses données au cours de la mission. Au moment de la rédaction de ce manuscrit, le banc d’étalonnage était en cours d’assemblage et j’ai donc pu conclure ce travail par la confrontation de mon modèle aux résultats expérimentaux de validation de certaines voies optiques du dispositif d’étalonnage. / The ESA JUICE mission will only be the third mission fully dedicated to exploring the Jupiter system, and the first with a specific focus on the icy Galilean moons that may harbor oceans of liquid water. Planned for launch in 2022 for a Jovian orbit insertion in late 2029, the probe will carry MAJIS among its 11 instruments, an imaging spectrometer operating from the visible to medium infrared wavelengths. This type of instrument provides very comprehensive data of the observed surface or atmosphere/exosphere: its high spatial resolution capability provides geomorphological information, such as the presence of craters or faults that mark the age and activity of the terrain, while for each pixel a spectrum is acquired. This spectrum, ranging from 0.5 to 5.5 $mu$m, yields physical and chemical information on the region of interest, thus placed in its geomorphic context. The Institut d'Astrophysique Spatiale, my PhD host laboratory, has a legacy of development of such instruments, prominently OMEGA aboard the 2003 Mars Express probe, of which MAJIS is the latest and current project. In this context, my work’s aim was to contribute to the definition and implementation of the instrument’s calibration: to achieve that I first had to understand its scientific objectives and the resulting instrumental requirements, as well as mastering the characteristics of MAJIS subsystems. As part of that process, I analyzed recent data of Io acquired with SPHERE, an integral field spectrometer on the VLT, which possesses similarities with the expected data products of MAJIS. Though this satellite is a minor objective of the JUICE mission, I had to understand the instrument itself in order to reduce its data and the spectra analysis required the development of a photometric model of a surface observation which I confronted to the reality and to previous studies. The identification of many systematic biases in these data and the quantification of its spatial and spectral detection limits allowed me to highlight the critical aspect in the upcoming calibration phase of MAJIS in order to get interpretable in-flight data. To reach this goal the knowledge of the subsystems of the instrument is also necessary because their behavior will condition the calibration scenario and all their parameters will not be measured again on this occasion. I have therefore characterized, using dedicated optical benches, the focal plane of the instrument and especially its HgCdTe CMOS infrared detector. I was able to measure its most common characteristics, such as its dark current, full-well capacity, quantum efficiency, persistence and readout noise. The knowledge of QE and full-well depth was incorporated into an end-to-end radiometric model of MAJIS, which I fed with the spectral radiance of different scientific targets, including modeled ionian surface flows. In turn, this allowed me to select sources and optical solutions suitable for calibration. Due to the intense radiation levels in the Jovian magnetosphere, the detector’s resilience to radiations also needed to be studied. I was able to perform three test campaigns on control detectors with sources of electrons, protons and high energy photons, which allowed me to show the overall very good resilience of the focal plane to permanent damages and to validate the foreseen transient effects reduction algorithms. These three approaches required that I develop a complete photometric model of the instrument and of its calibration setup which I used to discuss its design and submit test sequences for the calibration campaign. I have worked with our laboratory engineers and industrials to design then build the calibration setup with the light sources that will allow measurement of the spatial, spectral and radiometric responses of the instrument, required to interpret its data during the mission.

Purpose: This research will study how hyperspectral sensoring can assess the moisture content of lettuce by monitoring its growth in indoor vertical farming. Research questions: “What accuracy can be achieved when using hyperspectral sensoring for assessing the moisture content of lettuce leaves grown in vertical farming?” “How can vertical farming contribute to sustainability in conjunction with integration of NIR spectroscopy?” Methodology: This study is an experimental study with a deductive approach in which experiments have been performed using the hyperspectral technologies singlespot sensor and the hyperspectral camera Specim FX17 to collect spectral data. To analyze the data from the experiments two regression models were used and trained to make it possible to predict future moisture content values in lettuce. In order to get a better understanding and analyze the results from the experiments, a literature review was also conducted on how hyperspectral imaging has been applied to assess the quality of food products. Conclusion: The achieved accuracies were 58.24 % and 65.54 % for the PLS regression model and the Neural Network model respectively. Employing hyperspectral sensoring as a non-destructive technique to assess the quality of food products grown and harvested in vertical farming systems, contributes to sustainability from several aspects such as reducing food waste, minimizing costs and detecting different quality attributes that affect the food products. / Syfte: Syftet med denna studie är att undersöka hur hyperspektral avbildning kan användas för att bedöma fuktigheten i sallad genom att kontrollera hur den växer i vertikal odling inomhus. Frågeställningar: “Vilken noggrannhet kan uppnås vid användning av hyperspektral avbildning för att bedöma fukthalt hos salladsblad som odlas i vertikal odling?” “Hur kan vertikal odling bidra till hållbarhet i kombination med integration av NIR spectroscopy?”  Metod: Denna studie är en experimentell studie med en kvantitativ metod inom vilken en deduktiv ansats har tillämpats genom användning av de hyperspektrala teknologierna single-spot sensor och hyperspektralkameran Specim FX17 för insamling av spektral data. För att analysera datan från experimenten skapades och tränades två olika regressionsmodeller till att möjliggöra förutsägning av framtida värden av fukthalt i sallad. För att få en bättre förståelse för och kunna göra en bättre analys av resultaten från experimenten, utfördes även en litteraturöversikt på vad tidigare forskning om tillämpningen av hyperspektral avbildning för kvalitetssäkring av matprodukter har visat. Slutsats: Noggrannheten för PLS-regressionsmodellen var 58,24 % och 65,54 % för Neural Network-modellen. Minskat matsvinn och kostnader samt upptäcka olika kvalitetsattribut som påverkar livsmedelsprodukterna är de hållbara resultaten vid bedömning av kvalitet via hyperspektral sensing.

No description available.

No description available.

<p dir="ltr">The increasing global need for food security and sustainable agriculture underscores the urgency of advancing field phenotyping for enhanced plant breeding and crop management. Soybean, a major global protein source, is at the forefront of these advancements. Proximal sensing in soybean phenotyping offers a higher signal-to-noise ratio and resolution but has been underutilized in large-scale field applications due to low throughput and high labor costs. Moreover, there is an absence of automated solutions for in vivo proximal phenotyping of dicot plants. This thesis addresses these gaps by introducing a comprehensive, technologically sophisticated approach to modern field phenotyping.</p><p dir="ltr">Fully Automated Proximal Hyperspectral Imaging System: The first chapter presents the development of a cutting-edge hyperspectral imaging system integrated with a robotic arm. This system surpasses traditional imaging limitations, providing enhanced close-range data for accurate plant health assessment.</p><p dir="ltr">Robust Leaf Pose Estimation: The second chapter discusses the application of deep learning for accurate leaf pose estimation. This advancement is crucial for in-depth plant analysis, fostering better insights into plant health and growth, thereby contributing to increased crop yield and disease resistance.</p><p dir="ltr">PhenoBee – A Drone Mobility Platform: The third chapter introduces 'PhenoBee,' a dronebased platform designed for extensive field phenotyping. This innovative technology significantly broadens the capabilities of field data collection, showcasing its viability for widespread aerial phenotyping.</p><p dir="ltr">Adaptive Sampling for Dynamic Waypoint Planning: The final chapter details an adaptive sampling algorithm for efficient, real-time waypoint planning. This strategic approach enhances field scouting efficiency and precision, ensuring optimal data acquisition.</p><p dir="ltr">By integrating deep learning, robotic automation, aerial mobility, and intelligent sampling algorithms, the proposed solution revolutionizes the adaptation of in vivo proximal phenotyping on a large scale. The findings of this study highlight the potential to automate agriculture activities with high scalability and identify nutrient deficiencies, diseases, and chemical damage in crops earlier, thereby preventing yield loss, improving food quality, and expediting the development of agricultural products. Collectively, these advancements pave the way for more effective and efficient plant breeding and crop management, directly contributing to the enhancement of global food production systems. This study not only addresses current limitations in field phenotyping but also sets a new standard for technological innovation in agriculture.</p>

No description available.

Oberflächennahe Prozesse werden durch die dynamischen Eigenschaften der Bodenoberfläche besonders beeinflusst. Zwar sind die kausalen Zusammenhänge dieser Prozesse weitestgehend bekannt, doch gibt es einen Mangel an verfügbaren Datenquellen und Erhebungsmethoden, die es erlauben, die Prozesse auf unterschiedlichen Skalen zu quantifizieren. Das Ziel dieser Arbeit bestand darin, das Potential ausgewählter moderner Fernerkundungstechnologien zu bewerten, relevante Bodeneigenschaften zu quantifizieren und damit das Verständnis von oberflächennahen Prozessen in degradierten Landschaften zu verbessern. Das Studiengebiet befand sich in einer Rekultivierunglandschaft des Niederlausitzer Braunkohletagebaus Welzow-Süd. Die Größe von 4 ha ermöglichte eine umfassende, interdisziplinäre und multi-temporale Analyse der Bodeneigenschaften auf Grundlage von Fernerkundungsmethoden sowie hydrologischen und bodenkundlichen Feld- und Labormessungen. Die Quantifizierung der Bodenfeuchte als eine entscheidende Variable für Infiltrations- und Abflussprozesse war das Ziel von labor- und feldspektroskopischen Messungen sowie von hyperspektralen Flugzeugscanner-Messungen. Der hierbei entwickelte Normalized Soil Moisture Index (NSMI) wurde als optimales Quantifizierungsmodell für Oberflächen-Bodenfeuchte im Feld ermittelt. Bodenrauhigkeit wurde in hoher Präzision durch Anwendung eines stationären Laserscanners gemessen und in Form neuartiger multi-skalarer Indizes quantifiziert. Die Analyse der raum-zeitlichen Verteilungen ermöglichte die Identifizierung von Rauhigkeitsmustern, die unter dem Einfluss der Erosion im Feld entstanden. Diese Arbeit entwickelte neuartige Methoden und Indizes zur Quantifizierung von Oberflächen-Bodenfeuchte und Rauhigkeit im Feld. Für die Zukunft verspricht deren Anwendung die Entwicklung eines tieferen Verständnisses von Bodenerosionsprozessen sowie die Sammlung wertvoller Daten durch Monitoring- und Modellierungskampagnen. / Soil processes taking place in the context of erosion and land degradation are highly dependent on the properties of the surface. While the causes and effects of such processes are commonly well understood on a conceptual level, there is a lack of adequate data sources allowing for their quantification at various spatial scales. The main goal of this thesis was to assess the role of state-of-the-art remote sensing methods for the quantification of soil properties with the aim to improve the understanding of surface processes taking place in a degraded landscape. The chosen study area of 4 ha size located in a lignite mine in eastern Germany allowed for a comprehensive, interdisciplinary and multi-temporal analysis of surface properties based on remote sensing, pedological and hydrological measurements. The quantification of surface soil moisture as an important variable for infiltration and runoff processes has been the objective in laboratory and field spectroscopic experiments as well as in airborne hyperspectral measurements. The newly developed Normalized Soil Moisture Index (NSMI) was identified as the most robust quantifier for surface soil moisture in the field. Surface roughness was successfully quantified at high precision in form of novel multiscale indices derived from datasets collected with a stationary laser scanning device. The analysis of spatiotemporal roughness distributions allowed for the detection of distinct patterns that developed under the influence of soil erosion in the field. The thesis developed a set of methods and indices that successfully implement the quantification of surface soil moisture and roughness in the field. For the future, the application of these methods promises further insights into the details of soil erosion processes taking place as well as the collection of invaluable datasets to be used for soil erosion monitoring and modeling campaigns.

Cancer is a world menace. After years of endeavor seeking the end of it, people started to realize that no matter how powerful the therapy could be, detection at early stage is always a cheaper, easier and more successful solution compared with curative methods for cancer developed onto its advanced stage. However, relatively few early-detection approaches have proven sufficiently effective and practical for mass use as a point-of-care tool. An early-cancer screening tool integrating the desired features of sensitive, informative, portable, and cost-effective is in need for the doctors. The progress in optical imaging and Micro-electro-mechanical system (MEMS) technology offers a promise for an innovative cancer screening alternative that is non-invasive, radiation-free, portable and potentially cost-effective. This dissertation investigates handheld instrumentation as multi-modalities of miniature imaging probes with various designs of MEMS devices, to obtain real-time images of epithelial tissue optical and physiological properties, combining the quantitative advantages of spectral analysis with the qualitative benefits of imaging to distinguish early cancer. This dissertation in sequence presents the handheld instruments in the fashions of Laser-scanning confocal microscopy (LSCM), optical diffuse reflectance imaging, nonlinear optical imaging modalities with their subsequent image-guided managements in oral cancer, skin cancer detection, circulating tumor cell (CTC) imaging, and imaging guided surgeries. One of the main challenges facing miniaturization lies in the mechanism of beam deflection across the sample. This dissertation introduces two generations of MEMS devices desgined, fabricated and incorporated in the imaging probes. A two-axis vertical comb driven silicon micromirror was used in the development of a handheld LSCM for oral cancer detection. Though obtaining numerous advantages, this first generation silicon MEMS micromirror suffers from small aperture size and high voltage requirement for actuation, which result in low collection efficiency in fluorescence imaging and medial safety concerns, respectively. Therefore a stainless steel scanner compatible with electrical discharge machining (EDM) process was fabricated with simplified process, low-voltage magnetic actuation and large fluorescence collection efficiency, with its capability demonstrated in the incorporation and embodiment of a handheld hyperspectral nonlinear imaging probe. Besides, software and controlling innovations for handheld imaging modalities are presented. A feedback controlling system for MEMS scanning status monitoring was developed for stabilized imaging rendering. For the sake of further improved imaging stability in handheld imaging and to enable on-site mosaic for large field viewing, a handheld mosaic system was developed and presented. / text

For the first time in Lithuania the foliage spectral reflectance properties of common tree species were investigated using hyperspectral imaging. The methodological outline was formulated and the procedures of practical hyperspectral imaging application were developed to stimulate the progress of hyperspectral remote sensing in Lithuanian forestry. Information extracted from foliage hyperspectral reflectance data was used to accurately determine forest tree species and the provenances of Scots Pine trees. The satisfactory results of determination of Scots Pine crown defoliation and the concentration of some needles chemical constituents were achieved investigating the foliar hyperspectral reflectance, too. The first spectral libraries of common Lithuanian tree species foliar reflectance were built considering the growing season. / Suformuoti hiperspektrinio skenavimo naudojimo įvairioms miško medžių savybėms tirti metodiniai ir praktiniai pagrindai – sukurtos ir išbandytos mėginių paėmimo, jų paruošimo skenuoti, skenavimo atlikimo ir gautos informacijos apdorojimo metodikos, kurios aprobuotos vykdant mokslinius tyrimus. Nustatyti vegetacijos sezono momentai, kuriais skirtingų miško medžių rūšių atpažinimas nuotoliniu būdu pagal jų spektrinus atspindžius būtų tiksliausias, o tai sudaro prielaidas tobulinti kitas nuotoliniais. Pasiūlyti metodai paprastosios pušies spyglių kai kurių cheminių elementų koncentracijai nustatyti naudojant hiperspektrinį skenavimą. Sukurtos Lietuvos miškuose augančių pagrindinių medžių rūšių lapijos spektrinio atspindžio kreivių bibliotekos, naudotinos miškų inventorizacijoje, kalibruoti ir klasifikuoti orlaiviuose sumontuotais jutikliais išgautus Lietuvos medynų hiperspektrinius vaizdus.

Thesis (MSc Food Sc)--University of Stellenbosch, 2011. / ENGLISH ABSTRACT: The use of near infrared (NIR) hyperspectral imaging for distinguishing between pre-germinated and non pre-germinated barley, wheat and sorghum kernels and, the effect of kernel shape on hyperspectral images, have been investigated. Two sample sets were imaged. The first sample set was divided into six subsets; these subsets were treated with water and left to pre-germinate for different times (0, 6, 9, 12, 18 and 24 hrs). Subset viability was determined with the tetrazolium test. The second sample set was divided into seven subsets, treated with water and left to pre-germinate for 0, 3, 6, 9, 12, 18, 24 or 30 hrs. Individual kernel viability was determined with the tetrazolium test. NIR hyperspectral images were acquired using two different SisuCHEMA hyperspectral imaging systems. The first system acquired images with a 150 9m spatial resolution (first sample set) and the second system acquired images with a 30 9m spatial resolution (second sample set). Principal component analysis (PCA) was performed and a distinction between pre-germinated and non pre-germinated kernels was illustrated in PCA score images. Loading line plots showed that the main compounds contributing to spectral variation were starch, water and protein. These compounds were related to starch and protein hydrolysis. The distinction between pre-germinated and non pre-germinated kernels observed in the 30 9m spatial resolution images indicated NIR hyperspectral imaging was perhaps sensing incomplete endosperm degradation. Some kernels determined as pre-germinated by the tetrazolium test had the same chemical composition according to the score image as non pre-germinated kernels in the 30 9m spatial resolution images. A partial least squares discriminant analysis (PLS-DA) model with two classes (pre- germinated and non pre-germinated) was developed for each of the cultivars of the first sample set. The two classes were assigned in principal component (PC) 1 vs. PC 5 score plots. The model created for the barley cultivars resulted in excessive false positives and false negatives. The prediction results of wheat cultivars revealed that the model had a classification rate of 81% for the non pre-germinated class and 93% for the pre-germinated class. The sorghum prediction results revealed that the model correctly predicted 97% of the non pre-germinated class and 93% of the pre-germinated class. Two different PLS-DA models were developed for one image of each cultivar of the 30 9m spatial resolution images. The first model was developed by assigning each kernel in the score image and the second model was developed by assigning pixels in the score plot to either the pre- germinated or non pre-germinated class. Model 1 resulted in excessive false negatives. Model 2 resulted in excessive false positives. The differences between pre-germinated and non pre-germinated kernels were only observed in higher (PC 5 and 6) order PCs of the 150 9m spatial resolution images. The lower (PCs 1 to 4) order PCs (of each commodity) were subsequently examined with the aid of classification gradients. Kernel shape effects were observed in these PCs. The use of NIR hyperspectral imaging for distinguishing between pre-germinated and non pre-germinated grain kernels shows promise. / AFRIKAANSE OPSOMMING: Die gebruik van naby infrarooi (NIR) hiperspektrale beeld-analise is geëvalueer om onderskeid te tref tussen voor-ontkiemde en nie-voor-ontkiemde gars, koring en sorghum korrels. Die effek van korrelvorm op hiperspektrale beelde is ook geëvalueer. Die eerste stel graan-monsters is gebruik vir 150 9m ruimtelike resolusie beelde en die tweede stel is gebruik vir 30 9m ruimtelike resolusie beelde. Die eerste kultivar stel is verdeel in ses sub-stelle en met gedistilleerde water behandel vir 0, 6, 9, 12, 18 en 24 hr. Sub-stel lewensvatbaarheid is met die tetrazolium toets vasgestel. Elke kultivar in die tweede stel is in sewe sub-stelle verdeel en is vir 0, 3, 6, 9, 12, 18, 24 of 30 hr geïnkubeer. Individuele korrel lewensvatbaarheid is met die tetrazolium toets vasgestel. NIR hiperspektrale beelde is verkry deur gebruik te maak van twee verskillende SisuCHEMA kameras. Die verskillende kameras is gebruik om verskillende resolusie (30 en 150 9m ruimtelike resolusie) beelde te verkry. Hoofkomponent analise (HKA) is uitgevoer en ’n verskil tussen voor- ontkiemde en nie-voor-ontkiemde korrels is waargeneem in die 150 9m ruimtelike resolusie beelde. HK ladings stippe het water, stysel en proteïene uitgesonder as die verbindings wat bydrae het tot spektrale variasie. ’n Verskil tussen die voor-ontkiemde korrels en nie-voor-ontkiemde korrels is ook gesien vir die 30 9m ruimtelike resolusie beelde. Dit is egter ook waargeneem dat sommige korrels as voor-ontkiem bepaal is deur die tetrazolium toets, maar dié korrels het dieselfde chemiese samestelling volgens die punte beeld as nie-voor-ontkiemde korrels. Onvolledige endosperm hidrolise is ’n moontlike verduideliking vir die verskynsel. Die verbindings wat bygedra het tot die variasie is water, stysel en proteïene. ’n Parsiële kleinste kwadrate diskriminant analise (PKW-DA) model met twee klasse is ontwikkel vir elke kultivar van die 150 9m ruimtelike resolusie beelde. Die klasse is aangewys in the punte stip. Die model met die hoogste variasie in Y is gekies om die ander kultivars van dieselfde kommoditeit te voorspel. The PKW-DA resultate vir die gars kultivars het getoon dat die model vals positiewes en vals negatiewes opgelewer het. Die koring PKW-DA model het ’n klassifikasie koers van 81% vir die nie-voor-ontkiemde klasse en 93% vir die voor-ontkiemde klasse opgelewer. The PKW-DA resultate vir sorghum het getoon dat die model ’n klassifikasie koers van 97% vir die nie-voor-ontkiemde klasse en 93% vir die voor-ontkiemde klasse opgelewer. Twee verskillende PKW-DA modelle is ontwikkel vir elke beeld van elke kultivar van die 30 9m ruimtelike resolusie beelde. Die eerste model is ontwikkel deur elke korrel in die punte beeld aan te wys tot een van twee klasse en die tweede model is ontwikkel deur die beeldelemente in die punte stip tot een van twee klasse toe te skryf. Model 1 het vals negatiewes opgelewer en model 2 vals positiewes. Die verskille tussen die nie-voor-ontkiemde en voor-ontkiemde korrels is eers verduidelik in hoër orde HK van die 150 9m ruimtelike resolusie beelde. Die laer orde HK is dus ondersoek vir hul bydrae tot spektrale variasie met die hulp van klassifikasie gradiënte. Korrel vorm effekte is waargeneem. Die gebruik van NIR hiperspektrale beelding om onderskeid te tref tussen voor-ontkiemde en nie-voor-ontkiemde graan korrels, lyk belowend.