{"id":2678,"date":"2026-11-12T20:39:05","date_gmt":"2026-11-12T20:39:05","guid":{"rendered":"https:\/\/aa1.es\/?p=2678"},"modified":"2026-11-14T20:23:34","modified_gmt":"2026-11-14T20:23:34","slug":"analysis-and-determination-of-particle-size-2","status":"publish","type":"post","link":"https:\/\/aa1.es\/?p=2678","title":{"rendered":"Hyperspectral Imaging Analysis"},"content":{"rendered":"\n<div class=\"wp-block-uagb-info-box uagb-block-25555b74 uagb-infobox__content-wrap  uagb-infobox-icon-left-title uagb-infobox-left uagb-infobox-image-valign-top\"><div class=\"uagb-ifb-content\"><div class=\"uagb-ifb-left-title-image\"><div class=\"uagb-ifb-icon-wrap\"><svg xmlns=\"https:\/\/www.w3.org\/2000\/svg\" viewBox=\"0 0 512 512\"><path d=\"M0 256C0 114.6 114.6 0 256 0C397.4 0 512 114.6 512 256C512 397.4 397.4 512 256 512C114.6 512 0 397.4 0 256zM371.8 211.8C382.7 200.9 382.7 183.1 371.8 172.2C360.9 161.3 343.1 161.3 332.2 172.2L224 280.4L179.8 236.2C168.9 225.3 151.1 225.3 140.2 236.2C129.3 247.1 129.3 264.9 140.2 275.8L204.2 339.8C215.1 350.7 232.9 350.7 243.8 339.8L371.8 211.8z\"><\/path><\/svg><\/div><div class=\"uagb-ifb-title-wrap\"><h3 class=\"uagb-ifb-title\">SPECIM FX10 and FX17<\/h3><\/div><\/div><p class=\"uagb-ifb-desc\">Hyperspectral Imaging Analysis in the Vis-NIR and NIR Ranges<\/p><\/div><\/div>\n\n\n\n<div style=\"height:45px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<p class=\"wp-block-paragraph\">The SPECIM cameras allow the imaging of any kind of sample without previous preparation in both 400-1000 nm (FX10) and the 900-1700 nm (FX17) using a push-broom system. This allows the system to obtain spectral maps of various samples, enabling chemical analysis of specific regions of the samples. For this reason, hyperspectral imaging systems are ideal for:<\/p>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-a1e13a7c\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Run <strong>non-destructive, non-intrusive, contactless<\/strong> analysis.<\/li>\n\n\n\n<li>Produce <strong>distribution maps<\/strong> of parameters such as moisture, fat or sugar content.<\/li>\n\n\n\n<li><strong>Classify<\/strong> samples and <strong>detect<\/strong> defects, contaminants or foreign bodies.<\/li>\n<\/ul>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-a1e13a7c\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<p class=\"wp-block-paragraph\">Acquisition needs very little input from the user. The sample is scanned on a motorised stage under halogen illumination, and the instrument is calibrated against standard references before each measurement, so results can be reproduced from one session to the next.<\/p>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-11e43136\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<h5 class=\"wp-block-heading\"><strong>Spectral range<\/strong><\/h5>\n\n\n\n<div class=\"wp-block-uagb-info-box uagb-block-bf7f945a uagb-infobox__content-wrap  uagb-infobox-icon-left-title uagb-infobox-left uagb-infobox-image-valign-top wp-block-uagb-info-box--has-margin fondo-gris-claro has-custom-css wp-custom-css-c8cf60f0\"><div class=\"uagb-infobox-margin-wrapper\"><div class=\"uagb-ifb-content\"><div class=\"uagb-ifb-left-title-image\"><div class=\"uagb-ifb-icon-wrap\"><svg xmlns=\"https:\/\/www.w3.org\/2000\/svg\" viewBox=\"0 0 384 512\"><path d=\"M361 215C375.3 223.8 384 239.3 384 256C384 272.7 375.3 288.2 361 296.1L73.03 472.1C58.21 482 39.66 482.4 24.52 473.9C9.377 465.4 0 449.4 0 432V80C0 62.64 9.377 46.63 24.52 38.13C39.66 29.64 58.21 29.99 73.03 39.04L361 215z\"><\/path><\/svg><\/div><div class=\"uagb-ifb-title-wrap\"><h6 class=\"uagb-ifb-title\"><strong>SPECIM FX10 (Vis-NIR, 400\u20131000 nm):<\/strong><\/h6><\/div><\/div><p class=\"uagb-ifb-desc\"><br>CMOS sensor with 1024 spatial pixels.<br>Up to 224 spectral bands, with a spectral resolution of about 5.5 nm (FWHM).<br>Up to about 330 frames per second at full spectral range.<br>Sensitive to colour, pigments and surface features, and to the first water absorption bands.<\/p><\/div><\/div><\/div>\n\n\n\n<div class=\"wp-block-uagb-info-box uagb-block-1e735fe4 uagb-infobox__content-wrap  uagb-infobox-icon-left-title uagb-infobox-left uagb-infobox-image-valign-top wp-block-uagb-info-box--has-margin fondo-gris-claro has-custom-css wp-custom-css-d1ad6693\"><div class=\"uagb-infobox-margin-wrapper\"><div class=\"uagb-ifb-content\"><div class=\"uagb-ifb-left-title-image\"><div class=\"uagb-ifb-icon-wrap\"><svg xmlns=\"https:\/\/www.w3.org\/2000\/svg\" viewBox=\"0 0 384 512\"><path d=\"M361 215C375.3 223.8 384 239.3 384 256C384 272.7 375.3 288.2 361 296.1L73.03 472.1C58.21 482 39.66 482.4 24.52 473.9C9.377 465.4 0 449.4 0 432V80C0 62.64 9.377 46.63 24.52 38.13C39.66 29.64 58.21 29.99 73.03 39.04L361 215z\"><\/path><\/svg><\/div><div class=\"uagb-ifb-title-wrap\"><h6 class=\"uagb-ifb-title\"><strong>SPECIM FX17 (NIR, 900\u20131700 nm):<\/strong><\/h6><\/div><\/div><p class=\"uagb-ifb-desc\"><br>Thermoelectrically cooled InGaAs sensor with 640 spatial pixels.<br>Up to 224 spectral bands, with a spectral resolution of about 8 nm (FWHM).<br>Up to 670 frames per second at full spectral range.<br>Sensitive to the C-H, O-H and N-H bonds found in water, fats, sugars and proteins.<\/p><\/div><\/div><\/div>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-50e65383\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<h5 class=\"wp-block-heading\"><strong>What is a hyperspectral image?<\/strong><\/h5>\n\n\n\n<p class=\"wp-block-paragraph\">A conventional camera records three colors per pixel: red, green and blue. A hyperspectral camera records a <strong>complete spectrum for every pixel<\/strong>. This image can be seen as a collection of hundreds of photographs of the same scene, each taken at a different wavelength. Each pixel has an \u201cspectral fingerprint\u201d while each of these photographs shows the different regions where sample absorbs or reflects light of a certain wavelength.<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-large\" style=\"margin-top:0;margin-right:0;margin-bottom:0;margin-left:0\"><img fetchpriority=\"high\" decoding=\"async\" width=\"1024\" height=\"526\" src=\"https:\/\/aa1.es\/wp-content\/uploads\/2026\/11\/What-is-a-hyperspectral-image-1024x526.jpg\" alt=\"\" class=\"wp-image-2694\" srcset=\"https:\/\/aa1.es\/wp-content\/uploads\/2026\/11\/What-is-a-hyperspectral-image-1024x526.jpg 1024w, https:\/\/aa1.es\/wp-content\/uploads\/2026\/11\/What-is-a-hyperspectral-image-300x154.jpg 300w, https:\/\/aa1.es\/wp-content\/uploads\/2026\/11\/What-is-a-hyperspectral-image-768x394.jpg 768w, https:\/\/aa1.es\/wp-content\/uploads\/2026\/11\/What-is-a-hyperspectral-image.jpg 1032w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-beb47b8a\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<h5 class=\"wp-block-heading\"><strong>How is the image acquired?<\/strong><\/h5>\n\n\n\n<p class=\"wp-block-paragraph\">The FX cameras work in <em>push-broom<\/em> mode: they capture one line of the sample at a time while the sample moves beneath them on a motorised stage. The full image is built up line by line. The scanning is controlled with Specim\u2019s LUMO Scanner software.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Before each measurement, the system is calibrated with two references:<\/p>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-a1e13a7c\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<ul class=\"wp-block-list\">\n<li>A <strong>white reference<\/strong>, a surface with a known reflectance close to 100%.<\/li>\n\n\n\n<li>A <strong>dark reference<\/strong>, taken with the shutter closed to record the sensor\u2019s own noise.<\/li>\n<\/ul>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-cd2044b2\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<p class=\"wp-block-paragraph\">These references turn the raw signal into <strong>relative reflectance<\/strong>. This makes measurements comparable across days, across samples and between the two cameras.<\/p>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-1c5ed587\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<h5 class=\"wp-block-heading\"><strong>What chemical information can we obtain?<\/strong><\/h5>\n\n\n\n<p class=\"wp-block-paragraph\">In the near-infrared, light excites overtones and combination bands of molecular vibrations. Each type of chemical bond absorbs in typical regions of the spectrum. Water is a good example:<\/p>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-7b6e47f3\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<ul class=\"wp-block-list\">\n<li>It has an absorption band at about <strong>970 nm<\/strong> (second overtone of the O-H stretch), which the FX10 can detect.<\/li>\n\n\n\n<li>It has a much stronger band at about <strong>1450 nm<\/strong> (first overtone of the O-H stretch), which the FX17 can detect.<\/li>\n<\/ul>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-8fe99491\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<p class=\"wp-block-paragraph\">By following these bands pixel by pixel, we can see which areas of a fruit, a fillet or a grain hold more moisture.<\/p>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-9df942e6\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<h5 class=\"wp-block-heading\"><strong>How do we turn the images into results?<\/strong><\/h5>\n\n\n\n<p class=\"wp-block-paragraph\">A single hyperspectral image holds hundreds of thousands of strongly correlated spectra. To extract useful information, the spectra are first pre-processed: the background is removed and the spectra are corrected for scattering and baseline effects with SNV, MSC or Savitzky\u2013Golay derivatives. They are then modelled with the machine learning methods our group specialises in:<\/p>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-6db9869f\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Random Forest (RF):<\/strong> an ensemble of decision trees that ranks the importance of each wavelength, which helps in designing simpler multispectral systems.<\/li>\n\n\n\n<li><strong>Support Vector Machines (SVM):<\/strong> find the best boundary between classes, even with small datasets and non-linear relationships.<\/li>\n\n\n\n<li><strong>Artificial Neural Networks (ANN):<\/strong> learn complex relationships between the spectrum and the target property, for both classification and regression.<\/li>\n<\/ul>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-092d5ad4\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<p class=\"wp-block-paragraph\">These models work pixel by pixel, so the result is a classification or chemical map of the whole sample. To also use spatial context, <strong>Deep Learning<\/strong> models can be applied:<\/p>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-849afb67\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<div class=\"wp-block-uagb-info-box uagb-block-276bf1b2 uagb-infobox__content-wrap  uagb-infobox-icon-left-title uagb-infobox-left uagb-infobox-image-valign-top wp-block-uagb-info-box--has-margin fondo-gris-claro has-custom-css wp-custom-css-4787dac2\"><div class=\"uagb-infobox-margin-wrapper\"><div class=\"uagb-ifb-content\"><div class=\"uagb-ifb-left-title-image\"><div class=\"uagb-ifb-icon-wrap\"><svg xmlns=\"https:\/\/www.w3.org\/2000\/svg\" viewBox=\"0 0 384 512\"><path d=\"M361 215C375.3 223.8 384 239.3 384 256C384 272.7 375.3 288.2 361 296.1L73.03 472.1C58.21 482 39.66 482.4 24.52 473.9C9.377 465.4 0 449.4 0 432V80C0 62.64 9.377 46.63 24.52 38.13C39.66 29.64 58.21 29.99 73.03 39.04L361 215z\"><\/path><\/svg><\/div><div class=\"uagb-ifb-title-wrap\"><h6 class=\"uagb-ifb-title\"><strong><strong>U-Net<\/strong><\/strong>:<\/h6><\/div><\/div><p class=\"uagb-ifb-desc\"><br>A convolutional encoder\u2013decoder network that classifies every pixel in a single step, for example to delimit bruises, contaminated areas or foreign bodies.<\/p><\/div><\/div><\/div>\n\n\n\n<div class=\"wp-block-uagb-separator uagb-block-cbeaf4dc\"><div class=\"wp-block-uagb-separator__inner\" style=\"--my-background-image:\"><\/div><\/div>\n\n\n\n<h5 class=\"wp-block-heading\"><strong>Applications<\/strong><\/h5>\n\n\n\n<p class=\"wp-block-paragraph\">The system is used mainly in the agri-food sector, for example:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Quality control and ripeness assessment of fruits and vegetables, including sugar content.<\/li>\n\n\n\n<li>Detection of bruises, rot and fungal contamination.<\/li>\n\n\n\n<li>Moisture, fat and protein content in meat, fish, cereals, nuts and baked goods.<\/li>\n\n\n\n<li>Detection of food fraud and adulteration.<\/li>\n\n\n\n<li>Sorting of seeds, grains and raw materials.<\/li>\n\n\n\n<li>Identification of plastics and packaging materials.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n","protected":false},"excerpt":{"rendered":"<p>The SPECIM cameras allow the imaging of any kind of sample without previous preparation in both 400-1000 nm (FX10) and the 900-1700 nm (FX17) using a push-broom system. This allows the system to obtain spectral maps of various samples, enabling chemical analysis of specific regions of the samples. For this reason, hyperspectral imaging systems are ideal for:<\/p>\n","protected":false},"author":1,"featured_media":2844,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_uag_custom_page_level_css":"","footnotes":"","_members_access_role":[],"_members_access_error":""},"categories":[11],"tags":[],"class_list":["post-2678","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-servicio"],"uagb_featured_image_src":{"full":["https:\/\/aa1.es\/wp-content\/uploads\/2026\/11\/Camara-hiperespectral.jpg",1745,1099,false],"thumbnail":["https:\/\/aa1.es\/wp-content\/uploads\/2026\/11\/Camara-hiperespectral-150x150.jpg",150,150,true],"medium":["https:\/\/aa1.es\/wp-content\/uploads\/2026\/11\/Camara-hiperespectral-300x189.jpg",300,189,true],"medium_large":["https:\/\/aa1.es\/wp-content\/uploads\/2026\/11\/Camara-hiperespectral-768x484.jpg",768,484,true],"large":["https:\/\/aa1.es\/wp-content\/uploads\/2026\/11\/Camara-hiperespectral-1024x645.jpg",800,504,true],"1536x1536":["https:\/\/aa1.es\/wp-content\/uploads\/2026\/11\/Camara-hiperespectral-1536x967.jpg",1536,967,true],"2048x2048":["https:\/\/aa1.es\/wp-content\/uploads\/2026\/11\/Camara-hiperespectral.jpg",1745,1099,false]},"uagb_author_info":{"display_name":"AA1 Grupo de Investigaciones Agrarias y Alimentarias","author_link":"https:\/\/aa1.es\/?author=1"},"uagb_comment_info":0,"uagb_excerpt":"The SPECIM cameras allow the imaging of any kind of sample without previous preparation in both 400-1000 nm (FX10) and the 900-1700 nm (FX17) using a push-broom system. This allows the system to obtain spectral maps of various samples, enabling chemical analysis of specific regions of the samples. For this reason, hyperspectral imaging systems are&hellip;","_links":{"self":[{"href":"https:\/\/aa1.es\/index.php?rest_route=\/wp\/v2\/posts\/2678","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/aa1.es\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/aa1.es\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/aa1.es\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/aa1.es\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=2678"}],"version-history":[{"count":65,"href":"https:\/\/aa1.es\/index.php?rest_route=\/wp\/v2\/posts\/2678\/revisions"}],"predecessor-version":[{"id":2845,"href":"https:\/\/aa1.es\/index.php?rest_route=\/wp\/v2\/posts\/2678\/revisions\/2845"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/aa1.es\/index.php?rest_route=\/wp\/v2\/media\/2844"}],"wp:attachment":[{"href":"https:\/\/aa1.es\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=2678"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/aa1.es\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=2678"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/aa1.es\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=2678"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}