{"id":10617,"date":"2024-07-13T20:50:45","date_gmt":"2024-07-13T18:50:45","guid":{"rendered":"https:\/\/geopard.tech\/?p=10617"},"modified":"2024-07-13T20:50:51","modified_gmt":"2024-07-13T18:50:51","slug":"la-5g-a-permis-un-apprentissage-en-temps-reel-dans-lagriculture-durable-une-etude-sur-la-production-de-betteraves-sucrieres","status":"publish","type":"post","link":"https:\/\/geopard.tech\/fr\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/","title":{"rendered":"Apprentissage en temps r\u00e9el bas\u00e9 sur la technologie 5G dans l'agriculture durable : Une \u00e9tude sur la betterave \u00e0 sucre"},"content":{"rendered":"<p class=\"wp-block-paragraph\">Nous sommes ravis d&#039;annoncer la r\u00e9ussite du projet \u201c Les r\u00e9seaux 5G comme catalyseur de l&#039;apprentissage en temps r\u00e9el dans l&#039;agriculture durable \u201d, soutenu en partie par le minist\u00e8re des Affaires \u00e9conomiques, de l&#039;Industrie, de l&#039;Action climatique et de l&#039;\u00c9nergie du Land de Rh\u00e9nanie-du-Nord-Westphalie.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter\"><img decoding=\"async\" src=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2022\/08\/image-2.png?w=1620&amp;ssl=1\" alt=\"\"\/><\/figure>\n<\/div>\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Cette initiative repr\u00e9sente un pas en avant significatif dans l&#039;exploration du potentiel transformateur de la technologie 5G dans l&#039;agriculture, visant sp\u00e9cifiquement \u00e0 am\u00e9liorer les aspects \u00e9cologiques, \u00e9conomiques et durables de la culture de la betterave sucri\u00e8re.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Elle a tir\u00e9 parti de la faible latence de la 5G pour int\u00e9grer en temps r\u00e9el des syst\u00e8mes informatiques avanc\u00e9s, permettant des r\u00e9ponses imm\u00e9diates aux donn\u00e9es des capteurs et de positionnement dans des d\u00e9lais pr\u00e9d\u00e9finis.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img data-recalc-dims=\"1\" fetchpriority=\"high\" decoding=\"async\" width=\"624\" height=\"386\" data-attachment-id=\"10624\" data-permalink=\"https:\/\/geopard.tech\/fr\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/picture-from-the-final-event-of-the-project-presentation-at-hochschule-hamm-lippstadt-hshl\/\" data-orig-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Picture-from-the-final-event-of-the-project-presentation-at-Hochschule-Hamm-Lippstadt-HSHL.png?fit=624%2C386&amp;ssl=1\" data-orig-size=\"624,386\" data-comments-opened=\"1\" data-image-meta=\"{&quot;aperture&quot;:&quot;0&quot;,&quot;credit&quot;:&quot;&quot;,&quot;camera&quot;:&quot;&quot;,&quot;caption&quot;:&quot;&quot;,&quot;created_timestamp&quot;:&quot;0&quot;,&quot;copyright&quot;:&quot;&quot;,&quot;focal_length&quot;:&quot;0&quot;,&quot;iso&quot;:&quot;0&quot;,&quot;shutter_speed&quot;:&quot;0&quot;,&quot;title&quot;:&quot;&quot;,&quot;orientation&quot;:&quot;0&quot;}\" data-image-title=\"Picture from the final event of the project presentation at Hochschule Hamm-Lippstadt (HSHL)\" data-image-description=\"\" data-image-caption=\"\" data-large-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Picture-from-the-final-event-of-the-project-presentation-at-Hochschule-Hamm-Lippstadt-HSHL.png?fit=624%2C386&amp;ssl=1\" src=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Picture-from-the-final-event-of-the-project-presentation-at-Hochschule-Hamm-Lippstadt-HSHL.png?resize=624%2C386&#038;ssl=1\" alt=\"Photo prise lors de la pr\u00e9sentation finale du projet \u00e0 la Hochschule Hamm-Lippstadt (HSHL).\" class=\"wp-image-10624\" style=\"width:836px;height:auto\" srcset=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Picture-from-the-final-event-of-the-project-presentation-at-Hochschule-Hamm-Lippstadt-HSHL.png?w=624&amp;ssl=1 624w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Picture-from-the-final-event-of-the-project-presentation-at-Hochschule-Hamm-Lippstadt-HSHL.png?resize=300%2C186&amp;ssl=1 300w\" sizes=\"(max-width: 624px) 100vw, 624px\" \/><figcaption class=\"wp-element-caption\">Photo prise lors de la pr\u00e9sentation finale du projet \u00e0 la Hochschule Hamm-Lippstadt (HSHL).<\/figcaption><\/figure>\n<\/div>\n\n\n<h2 class=\"wp-block-heading\" id=\"h-project-focus-and-partnership\">Objectifs et partenariats du projet<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">En collaboration avec des partenaires du HSHL et avec le soutien de Pfeifer &amp; Langen, le projet s&#039;est concentr\u00e9 sur l&#039;\u00e9tude du cycle de vie complet de la culture de la betterave sucri\u00e8re sur les exploitations des partenaires. Il visait \u00e0 d\u00e9montrer comment la 5G pourrait jouer un r\u00f4le de catalyseur technologique essentiel au sein du secteur agricole de Rh\u00e9nanie-du-Nord-Westphalie, en mettant en \u00e9vidence son potentiel en tant que facteur d&#039;innovation et d&#039;efficacit\u00e9.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"h-role-of-geopard-agriculture\">R\u00f4le de l&#039;agriculture GeoPard<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">GeoPard Agriculture a jou\u00e9 un r\u00f4le crucial dans la d\u00e9finition et la mise en \u0153uvre d&#039;aspects cl\u00e9s du projet, notamment les sc\u00e9narios de d\u00e9tection, de surveillance et de pr\u00e9vision de la production des plantes. Nous avons d\u00e9velopp\u00e9 un prototype de syst\u00e8me d&#039;IA adapt\u00e9 \u00e0 l&#039;environnement agricole 5G, ex\u00e9cut\u00e9 des mod\u00e8les au sein d&#039;une infrastructure cloud et cr\u00e9\u00e9 une application mobile permettant une interaction en temps r\u00e9el avec ces mod\u00e8les.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"h-technological-integration\">Int\u00e9gration technologique<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Des m\u00e9thodes d&#039;intelligence artificielle (IA) ont \u00e9t\u00e9 d\u00e9ploy\u00e9es via une infrastructure cloud robuste dot\u00e9e de capacit\u00e9s de calcul \u00e9lev\u00e9es. Les algorithmes d&#039;IA ont cat\u00e9goris\u00e9 les plantes en temps r\u00e9el lors de chaque croisement et ont suivi leur croissance tout au long de leur cycle de vie, \u00e9liminant ainsi le besoin de visites de terrain inutiles \u00e0 des fins de simple collecte de donn\u00e9es.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Cette avanc\u00e9e a permis l&#039;application pr\u00e9cise d&#039;engrais et de produits phytosanitaires, en ajustant dynamiquement les doses d&#039;application lors des croisements gr\u00e2ce \u00e0 des algorithmes d&#039;apprentissage automatique.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"h-deployment-of-unmanned-vehicles\">D\u00e9ploiement de v\u00e9hicules sans pilote<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">De plus, le projet a tir\u00e9 parti de la faible latence de la 5G pour d\u00e9ployer des v\u00e9hicules autonomes d\u00e9di\u00e9s \u00e0 la surveillance des cultures et \u00e0 la collecte de donn\u00e9es. Ces v\u00e9hicules ont jou\u00e9 un r\u00f4le crucial dans l&#039;obtention d&#039;informations en temps r\u00e9el et dans l&#039;optimisation des pratiques agricoles.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"h-project-outcomes-enhancing-sugar-beet-production-with-5g-technology\">R\u00e9sultats du projet : Am\u00e9lioration de la production de betteraves sucri\u00e8res gr\u00e2ce \u00e0 la technologie 5G<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Le projet a d\u00e9montr\u00e9 comment la technologie 5G pouvait transformer en profondeur le secteur agricole de Rh\u00e9nanie-du-Nord-Westphalie en analysant l&#039;ensemble du cycle de vie de la culture de la betterave sucri\u00e8re et en mettant en \u00e9vidence les am\u00e9liorations substantielles permises par cette technologie. Toutefois, afin de pr\u00e9senter efficacement les r\u00e9sultats du projet, les chercheurs ont utilis\u00e9 des lots de travaux comprenant diff\u00e9rents sc\u00e9narios et infrastructures.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full\"><img data-recalc-dims=\"1\" decoding=\"async\" width=\"810\" height=\"610\" data-attachment-id=\"10625\" data-permalink=\"https:\/\/geopard.tech\/fr\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/sugar-beet-test-field\/\" data-orig-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-test-field.png?fit=2048%2C1542&amp;ssl=1\" data-orig-size=\"2048,1542\" data-comments-opened=\"1\" data-image-meta=\"{&quot;aperture&quot;:&quot;0&quot;,&quot;credit&quot;:&quot;&quot;,&quot;camera&quot;:&quot;&quot;,&quot;caption&quot;:&quot;&quot;,&quot;created_timestamp&quot;:&quot;0&quot;,&quot;copyright&quot;:&quot;&quot;,&quot;focal_length&quot;:&quot;0&quot;,&quot;iso&quot;:&quot;0&quot;,&quot;shutter_speed&quot;:&quot;0&quot;,&quot;title&quot;:&quot;&quot;,&quot;orientation&quot;:&quot;0&quot;}\" data-image-title=\"Sugar beet test field\" data-image-description=\"\" data-image-caption=\"\" data-large-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-test-field.png?fit=1024%2C771&amp;ssl=1\" src=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-test-field.png?resize=810%2C610&#038;ssl=1\" alt=\"champ d&#039;essai de betteraves sucri\u00e8res \" class=\"wp-image-10625\" srcset=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-test-field.png?w=2048&amp;ssl=1 2048w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-test-field.png?resize=300%2C226&amp;ssl=1 300w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-test-field.png?resize=1024%2C771&amp;ssl=1 1024w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-test-field.png?resize=768%2C578&amp;ssl=1 768w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-test-field.png?resize=1536%2C1157&amp;ssl=1 1536w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-test-field.png?resize=400%2C300&amp;ssl=1 400w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-test-field.png?resize=200%2C150&amp;ssl=1 200w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-test-field.png?resize=1200%2C904&amp;ssl=1 1200w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-test-field.png?w=1620&amp;ssl=1 1620w\" sizes=\"(max-width: 810px) 100vw, 810px\" \/><figcaption class=\"wp-element-caption\">champ d&#039;essai de betteraves sucri\u00e8res<\/figcaption><\/figure>\n<\/div>\n\n\n<h3 class=\"wp-block-heading\" id=\"h-scenario-definition-considering-existing-geodata-and-ml-infrastructure\">D\u00e9finition du sc\u00e9nario prenant en compte l&#039;infrastructure de g\u00e9odonn\u00e9es et d&#039;apprentissage automatique existante<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Le projet a d\u00e9montr\u00e9 comment les processus traditionnels du cycle de vie de la production de betteraves sucri\u00e8res pouvaient \u00eatre am\u00e9lior\u00e9s gr\u00e2ce \u00e0 l&#039;int\u00e9gration de la technologie 5G. Les principaux objectifs \u00e9taient les suivants\u00a0:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u00c9laboration de sc\u00e9narios pr\u00eats \u00e0 l&#039;emploi pour la reconnaissance, la surveillance et la pr\u00e9vision de la production des plantes.<\/li>\n\n\n\n<li>Exigences techniques \u00e9tablies n\u00e9cessaires au d\u00e9ploiement r\u00e9ussi de ces sc\u00e9narios.<\/li>\n\n\n\n<li>Identification et \u00e9valuation des indicateurs \u00e9cologiques et \u00e9conomiques pertinents afin d&#039;\u00e9valuer la valeur ajout\u00e9e apport\u00e9e par le r\u00e9seau 5G.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Cette phase a soulign\u00e9 l&#039;engagement du projet \u00e0 int\u00e9grer des technologies de pointe aux pratiques agricoles existantes. Cette architecture a tir\u00e9 parti de la connectivit\u00e9 haut d\u00e9bit du r\u00e9seau 5G pour faciliter la collecte et le traitement des donn\u00e9es en temps r\u00e9el entre les dispositifs p\u00e9riph\u00e9riques et le cloud. L&#039;infrastructure cloud a fourni les ressources essentielles \u00e0 l&#039;entra\u00eenement et au d\u00e9ploiement de mod\u00e8les d&#039;IA \u00e0 grande \u00e9chelle, tandis que la plateforme d&#039;IA a offert des outils performants pour le d\u00e9veloppement et le d\u00e9ploiement de ces mod\u00e8les. La couche applicative a pr\u00e9sent\u00e9 aux utilisateurs finaux des informations exploitables issues des mod\u00e8les d&#039;IA, am\u00e9liorant ainsi leurs capacit\u00e9s de prise de d\u00e9cision.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"h-machine-learning-and-ai-in-the-context-of-5g\">Apprentissage automatique et intelligence artificielle dans le contexte de la 5G<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Cette partie visait \u00e0 adapter les syst\u00e8mes d&#039;apprentissage automatique et d&#039;IA existants aux sc\u00e9narios d\u00e9crits ci-dessus, en les optimisant en cons\u00e9quence. Les principaux objectifs \u00e9taient les suivants\u00a0:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>D\u00e9finir les objectifs du syst\u00e8me et d\u00e9velopper son architecture.<\/li>\n\n\n\n<li>Collecte de donn\u00e9es de r\u00e9f\u00e9rence pour l&#039;entra\u00eenement et la validation des mod\u00e8les d&#039;IA.<\/li>\n\n\n\n<li>Cr\u00e9ation et annotation d&#039;une base de donn\u00e9es adapt\u00e9e \u00e0 l&#039;identification et au suivi des plantes.<\/li>\n\n\n\n<li>Int\u00e9gration transparente des mod\u00e8les d&#039;IA \u00e0 l&#039;infrastructure du r\u00e9seau 5G.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Dans cette phase, les dispositifs p\u00e9riph\u00e9riques \u00e9quip\u00e9s de cartes SIM mobiles utilisant la technologie 5G ont jou\u00e9 un r\u00f4le crucial. Les indicateurs cl\u00e9s de performance (KPI), tels que la latence et la latence de bout en bout (E2E), ont \u00e9t\u00e9 surveill\u00e9s de pr\u00e8s. Les mesures ont consist\u00e9 \u00e0 \u00e9valuer avec pr\u00e9cision la fiabilit\u00e9 et la disponibilit\u00e9 des paquets de donn\u00e9es re\u00e7us, ainsi qu&#039;\u00e0 analyser les d\u00e9bits de donn\u00e9es des utilisateurs et les d\u00e9bits de pointe.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">De plus, des hypoth\u00e8ses ont \u00e9t\u00e9 formul\u00e9es \u00e0 partir de la diffusion de vid\u00e9os en r\u00e9solution UHD au format MP4, transmises via le protocole TCP. Parmi les solutions envisag\u00e9es figuraient l&#039;optimisation sur des images uniques plut\u00f4t que sur des flux vid\u00e9o continus, l&#039;ex\u00e9cution d&#039;optimisations de base directement sur les p\u00e9riph\u00e9riques et la mise en \u0153uvre de techniques de quantification du mod\u00e8le pour am\u00e9liorer l&#039;efficacit\u00e9.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"h-cloud-infrastructure-and-aws-services\">Infrastructure cloud et services AWS<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Le projet s&#039;appuyait fortement sur une infrastructure cloud tirant parti des services AWS tels que Lambda, SageMaker, S3, CloudWatch et RDS, qui ont jou\u00e9 un r\u00f4le essentiel en fournissant les ressources n\u00e9cessaires \u00e0 l&#039;entra\u00eenement et au d\u00e9ploiement des mod\u00e8les d&#039;IA.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">AWS Lambda a \u00e9t\u00e9 utilis\u00e9 pour une gestion efficace des instances et la diffusion des applications, tandis qu&#039;AWS SageMaker a facilit\u00e9 la construction de pipelines d&#039;apprentissage automatique robustes. Des solutions de stockage telles que S3, CloudWatch et RDS \u00e9taient essentielles pour stocker les ensembles de donn\u00e9es et les journaux indispensables au fonctionnement des mod\u00e8les d&#039;apprentissage automatique et des r\u00e9seaux neuronaux.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full\"><img data-recalc-dims=\"1\" decoding=\"async\" width=\"800\" height=\"490\" data-attachment-id=\"10627\" data-permalink=\"https:\/\/geopard.tech\/fr\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/aws-cloud-infrastructure\/\" data-orig-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/AWS-cloud-infrastructure.png?fit=800%2C490&amp;ssl=1\" data-orig-size=\"800,490\" data-comments-opened=\"1\" data-image-meta=\"{&quot;aperture&quot;:&quot;0&quot;,&quot;credit&quot;:&quot;&quot;,&quot;camera&quot;:&quot;&quot;,&quot;caption&quot;:&quot;&quot;,&quot;created_timestamp&quot;:&quot;0&quot;,&quot;copyright&quot;:&quot;&quot;,&quot;focal_length&quot;:&quot;0&quot;,&quot;iso&quot;:&quot;0&quot;,&quot;shutter_speed&quot;:&quot;0&quot;,&quot;title&quot;:&quot;&quot;,&quot;orientation&quot;:&quot;0&quot;}\" data-image-title=\"AWS cloud infrastructure\" data-image-description=\"\" data-image-caption=\"\" data-large-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/AWS-cloud-infrastructure.png?fit=800%2C490&amp;ssl=1\" src=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/AWS-cloud-infrastructure.png?resize=800%2C490&#038;ssl=1\" alt=\"Infrastructure cloud AWS\" class=\"wp-image-10627\" srcset=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/AWS-cloud-infrastructure.png?w=800&amp;ssl=1 800w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/AWS-cloud-infrastructure.png?resize=300%2C184&amp;ssl=1 300w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/AWS-cloud-infrastructure.png?resize=768%2C470&amp;ssl=1 768w\" sizes=\"(max-width: 800px) 100vw, 800px\" \/><figcaption class=\"wp-element-caption\">Infrastructure cloud AWS<\/figcaption><\/figure>\n<\/div>\n\n\n<p class=\"wp-block-paragraph\">Par cons\u00e9quent, cette infrastructure a permis de soutenir les capacit\u00e9s de traitement des donn\u00e9es en temps r\u00e9el offertes par le r\u00e9seau 5G.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"h-5g-network-latency\">Latence du r\u00e9seau 5G<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Les r\u00e9seaux 5G ont \u00e9t\u00e9 con\u00e7us pour atteindre une latence ultra-faible, g\u00e9n\u00e9ralement comprise entre 1 et 10 millisecondes. Cette latence correspond au temps n\u00e9cessaire aux donn\u00e9es pour transiter entre les appareils mobiles et les serveurs AWS via le r\u00e9seau 5G. Les capacit\u00e9s de traitement propres \u00e0 chaque appareil, comme la vitesse de capture et de traitement des photos sur les smartphones dot\u00e9s de processeurs hautes performances, influent \u00e9galement sur la latence.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Les d\u00e9bits de chargement sur le r\u00e9seau 5G et la taille des photos ont influenc\u00e9 les temps de transfert des donn\u00e9es vers AWS. AWS a \u00e9galement contribu\u00e9 \u00e0 la latence avec les temps de traitement de t\u00e2ches telles que la d\u00e9tection et la segmentation par r\u00e9seau neuronal, qui variaient selon la complexit\u00e9 de l&#039;algorithme et l&#039;efficacit\u00e9 du service AWS. Apr\u00e8s traitement, les r\u00e9sultats ont \u00e9t\u00e9 t\u00e9l\u00e9charg\u00e9s sur les appareils mobiles, le temps de t\u00e9l\u00e9chargement \u00e9tant influenc\u00e9 par le d\u00e9bit 5G et la taille des donn\u00e9es.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"h-plant-recognition-using-ai\">Reconnaissance des plantes par l&#039;IA<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Dans le domaine de la reconnaissance des plantes, les processus bas\u00e9s sur l&#039;IA ont consist\u00e9 \u00e0 cr\u00e9er une base de donn\u00e9es exhaustive d&#039;images de plantes pour l&#039;entra\u00eenement d&#039;algorithmes fond\u00e9s sur des r\u00e9seaux neuronaux. Ces algorithmes ont \u00e9t\u00e9 entra\u00een\u00e9s \u00e0 distinguer les esp\u00e8ces de betterave sucri\u00e8re des autres plantes en reconnaissant des caract\u00e9ristiques sp\u00e9cifiques \u00e0 chaque type de plante, telles que la forme des feuilles, la couleur des fleurs, etc.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" width=\"810\" height=\"264\" data-attachment-id=\"10628\" data-permalink=\"https:\/\/geopard.tech\/fr\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/phenological-development-of-sugar-beet-plants\/\" data-orig-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Phenological-development-of-sugar-beet-plants.jpg?fit=2048%2C667&amp;ssl=1\" data-orig-size=\"2048,667\" data-comments-opened=\"1\" data-image-meta=\"{&quot;aperture&quot;:&quot;0&quot;,&quot;credit&quot;:&quot;&quot;,&quot;camera&quot;:&quot;&quot;,&quot;caption&quot;:&quot;&quot;,&quot;created_timestamp&quot;:&quot;0&quot;,&quot;copyright&quot;:&quot;&quot;,&quot;focal_length&quot;:&quot;0&quot;,&quot;iso&quot;:&quot;0&quot;,&quot;shutter_speed&quot;:&quot;0&quot;,&quot;title&quot;:&quot;&quot;,&quot;orientation&quot;:&quot;0&quot;}\" data-image-title=\"Phenological development of sugar beet plants\" data-image-description=\"\" data-image-caption=\"\" data-large-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Phenological-development-of-sugar-beet-plants.jpg?fit=1024%2C334&amp;ssl=1\" src=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Phenological-development-of-sugar-beet-plants.jpg?resize=810%2C264&#038;ssl=1\" alt=\"D\u00e9veloppement ph\u00e9nologique des plants de betterave sucri\u00e8re\" class=\"wp-image-10628\" srcset=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Phenological-development-of-sugar-beet-plants.jpg?w=2048&amp;ssl=1 2048w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Phenological-development-of-sugar-beet-plants.jpg?resize=300%2C98&amp;ssl=1 300w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Phenological-development-of-sugar-beet-plants.jpg?resize=1024%2C334&amp;ssl=1 1024w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Phenological-development-of-sugar-beet-plants.jpg?resize=768%2C250&amp;ssl=1 768w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Phenological-development-of-sugar-beet-plants.jpg?resize=1536%2C500&amp;ssl=1 1536w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Phenological-development-of-sugar-beet-plants.jpg?resize=1200%2C391&amp;ssl=1 1200w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Phenological-development-of-sugar-beet-plants.jpg?w=1620&amp;ssl=1 1620w\" sizes=\"(max-width: 810px) 100vw, 810px\" \/><figcaption class=\"wp-element-caption\">D\u00e9veloppement ph\u00e9nologique des plants de betterave sucri\u00e8re. Source\u00a0: <a href=\"https:\/\/www.mdpi.com\/2073-4395\/11\/7\/1277\" rel=\"nofollow\">https:\/\/www.mdpi.com\/2073-4395\/11\/7\/1277<\/a><\/figcaption><\/figure>\n<\/div>\n\n\n<p class=\"wp-block-paragraph\">Ici, par reconnaissance des plantes, nous entendons la t\u00e2che de d\u00e9tection des mauvaises herbes et de segmentation des plants de betterave sucri\u00e8re.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>D\u00e9tection des mauvaises herbes<\/strong><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Pour la d\u00e9tection des mauvaises herbes, le projet a utilis\u00e9 MobileNet-v3, entra\u00een\u00e9 gr\u00e2ce \u00e0 un important enrichissement des donn\u00e9es et \u00e0 un \u00e9chantillonnage pond\u00e9r\u00e9. Ce mod\u00e8le a atteint une pr\u00e9cision remarquable de 0,984 et une aire sous la courbe (AUC) de 0,998.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Segmentation de la betterave sucri\u00e8re<\/strong><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Pour les t\u00e2ches de segmentation, des mod\u00e8les tels que YOLACT, ResNeSt, SOLO et U-Net ont \u00e9t\u00e9 utilis\u00e9s afin de d\u00e9limiter pr\u00e9cis\u00e9ment les \u00e9chantillons individuels de betterave sucri\u00e8re dans les images. Le mod\u00e8le le plus performant a ensuite \u00e9t\u00e9 s\u00e9lectionn\u00e9 selon diff\u00e9rents crit\u00e8res\u00a0: vitesse, temps d\u2019inf\u00e9rence, etc. Les donn\u00e9es de segmentation provenaient d\u2019images RVB acquises par drone, redimensionn\u00e9es et annot\u00e9es \u00e0 des fins d\u2019entra\u00eenement et de validation.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Les t\u00e2ches de segmentation consistaient \u00e0 cr\u00e9er des masques d\u00e9limitant pr\u00e9cis\u00e9ment les contours des plantes. Cette m\u00e9thode a permis de r\u00e9duire l&#039;effort d&#039;annotation humaine tout en optimisant l&#039;efficacit\u00e9. En priorisant l&#039;\u00e9tiquetage des \u00e9chantillons complexes, les performances du mod\u00e8le ont \u00e9t\u00e9 consid\u00e9rablement am\u00e9lior\u00e9es. Les strat\u00e9gies de r\u00e9entra\u00eenement it\u00e9ratif et d&#039;\u00e9chantillonnage de l&#039;incertitude se sont av\u00e9r\u00e9es efficaces, atteignant des taux de pr\u00e9cision de segmentation sup\u00e9rieurs \u00e0 98% \u00e0 diff\u00e9rents stades de croissance.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" width=\"355\" height=\"146\" data-attachment-id=\"10629\" data-permalink=\"https:\/\/geopard.tech\/fr\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/example-of-input-output-of-segmentation\/\" data-orig-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Example-of-input-output-of-segmentation.png?fit=355%2C146&amp;ssl=1\" data-orig-size=\"355,146\" data-comments-opened=\"1\" data-image-meta=\"{&quot;aperture&quot;:&quot;0&quot;,&quot;credit&quot;:&quot;&quot;,&quot;camera&quot;:&quot;&quot;,&quot;caption&quot;:&quot;&quot;,&quot;created_timestamp&quot;:&quot;0&quot;,&quot;copyright&quot;:&quot;&quot;,&quot;focal_length&quot;:&quot;0&quot;,&quot;iso&quot;:&quot;0&quot;,&quot;shutter_speed&quot;:&quot;0&quot;,&quot;title&quot;:&quot;&quot;,&quot;orientation&quot;:&quot;0&quot;}\" data-image-title=\"Example of input-output of segmentation\" data-image-description=\"\" data-image-caption=\"\" data-large-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Example-of-input-output-of-segmentation.png?fit=355%2C146&amp;ssl=1\" src=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Example-of-input-output-of-segmentation.png?resize=355%2C146&#038;ssl=1\" alt=\"Exemple d&#039;entr\u00e9e-sortie de segmentation\" class=\"wp-image-10629\" style=\"width:767px;height:auto\" srcset=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Example-of-input-output-of-segmentation.png?w=355&amp;ssl=1 355w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Example-of-input-output-of-segmentation.png?resize=300%2C123&amp;ssl=1 300w\" sizes=\"(max-width: 355px) 100vw, 355px\" \/><figcaption class=\"wp-element-caption\">Exemple d&#039;entr\u00e9e-sortie de segmentation<\/figcaption><\/figure>\n<\/div>\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u00c9valuation du mod\u00e8le<\/strong><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Le mod\u00e8le a \u00e9t\u00e9 entra\u00een\u00e9 avec des augmentations de donn\u00e9es rigoureuses. Son \u00e9valuation a \u00e9t\u00e9 r\u00e9alis\u00e9e \u00e0 l&#039;aide de diff\u00e9rentes m\u00e9triques, dont l&#039;intersection sur l&#039;union (IoU). L&#039;analyse d&#039;inf\u00e9rence du mod\u00e8le construit, men\u00e9e sur un sous-ensemble du jeu de donn\u00e9es \u2018\u00a0plant seeds v2\u00a0\u2019, a d\u00e9montr\u00e9 une pr\u00e9cision de 81%. Le temps de calcul de l&#039;inf\u00e9rence \u00e9tait d&#039;environ 320 millisecondes apr\u00e8s une p\u00e9riode d&#039;initialisation de 7 secondes, n\u00e9cessaire une seule fois par session.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Dans le cadre de la surveillance des plantes par intelligence artificielle (IA), des cam\u00e9ras et des capteurs ont permis de recueillir des donn\u00e9es essentielles, analys\u00e9es ensuite par des algorithmes d&#039;apprentissage automatique et d&#039;IA. Cette analyse a jou\u00e9 un r\u00f4le crucial dans l&#039;\u00e9valuation de la sant\u00e9 des plantes, la d\u00e9tection du stress, des maladies ou d&#039;autres facteurs ayant un impact sur leur croissance.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Les applications s&#039;\u00e9tendent de l&#039;optimisation de la productivit\u00e9 agricole \u00e0 la surveillance des \u00e9cosyst\u00e8mes naturels comme les for\u00eats, en passant par le soutien aux efforts de conservation et l&#039;am\u00e9lioration de la compr\u00e9hension des impacts environnementaux.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"h-object-detection-in-plant-monitoring\">D\u00e9tection d&#039;objets dans la surveillance des installations<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Apr\u00e8s la segmentation des plants de betterave sucri\u00e8re, l&#039;\u00e9tape suivante consiste en la d\u00e9tection d&#039;objets afin de comprendre les sp\u00e9cificit\u00e9s de chaque plant en termes de sant\u00e9, de croissance et d&#039;autres facteurs. Pour la d\u00e9tection d&#039;objets dans le cadre de la surveillance des plants, des mod\u00e8les avanc\u00e9s tels que YOLOv4, MobileNetV2 et VGG-19, int\u00e9grant des m\u00e9canismes d&#039;attention, ont \u00e9t\u00e9 utilis\u00e9s. Ces mod\u00e8les ont analys\u00e9 des images segment\u00e9es de betteraves sucri\u00e8res pour d\u00e9tecter les zones de stress et de maladie, permettant ainsi des interventions pr\u00e9cises et cibl\u00e9es.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Le projet a franchi des \u00e9tapes importantes dans la d\u00e9tection des maladies, gr\u00e2ce \u00e0 l&#039;entra\u00eenement des mod\u00e8les ResNet-18 et ResNet-34 pr\u00e9-entra\u00een\u00e9s sur ImageNet. Ces mod\u00e8les ont d\u00e9montr\u00e9 une pr\u00e9cision remarquable de 0,88 dans l&#039;identification des maladies affectant les plants de betterave sucri\u00e8re, avec une aire sous la courbe ROC (AUC) de 0,898. Ils ont affich\u00e9 une grande fiabilit\u00e9 de pr\u00e9diction, distinguant avec pr\u00e9cision les plants malades des plants sains.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" width=\"414\" height=\"194\" data-attachment-id=\"10630\" data-permalink=\"https:\/\/geopard.tech\/fr\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/example-of-input-output-of-object-detection\/\" data-orig-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Example-of-input-output-of-object-detection.png?fit=414%2C194&amp;ssl=1\" data-orig-size=\"414,194\" data-comments-opened=\"1\" data-image-meta=\"{&quot;aperture&quot;:&quot;0&quot;,&quot;credit&quot;:&quot;&quot;,&quot;camera&quot;:&quot;&quot;,&quot;caption&quot;:&quot;&quot;,&quot;created_timestamp&quot;:&quot;0&quot;,&quot;copyright&quot;:&quot;&quot;,&quot;focal_length&quot;:&quot;0&quot;,&quot;iso&quot;:&quot;0&quot;,&quot;shutter_speed&quot;:&quot;0&quot;,&quot;title&quot;:&quot;&quot;,&quot;orientation&quot;:&quot;0&quot;}\" data-image-title=\"Example of input-output of object detection\" data-image-description=\"\" data-image-caption=\"\" data-large-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Example-of-input-output-of-object-detection.png?fit=414%2C194&amp;ssl=1\" src=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Example-of-input-output-of-object-detection.png?resize=414%2C194&#038;ssl=1\" alt=\"Exemple d&#039;entr\u00e9e-sortie de la d\u00e9tection d&#039;objets\" class=\"wp-image-10630\" style=\"width:750px;height:auto\" srcset=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Example-of-input-output-of-object-detection.png?w=414&amp;ssl=1 414w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Example-of-input-output-of-object-detection.png?resize=300%2C141&amp;ssl=1 300w\" sizes=\"(max-width: 414px) 100vw, 414px\" \/><figcaption class=\"wp-element-caption\">Exemple d&#039;entr\u00e9e-sortie de la d\u00e9tection d&#039;objets<\/figcaption><\/figure>\n<\/div>\n\n\n<p class=\"wp-block-paragraph\">Le projet a mis en \u0153uvre une approche syst\u00e9matique de d\u00e9tection des maladies, en segmentant les images en zones standardis\u00e9es. Ces zones ont ensuite fait l&#039;objet d&#039;une annotation minutieuse \u00e0 l&#039;aide d&#039;outils interactifs afin de localiser pr\u00e9cis\u00e9ment les zones affect\u00e9es par les maladies. La d\u00e9tection d&#039;objets a permis d&#039;am\u00e9liorer encore la pr\u00e9cision en d\u00e9limitant des cadres autour des plantes, facilitant ainsi un suivi pr\u00e9cis de leur \u00e9tat sanitaire.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"h-plant-production-prediction\">Pr\u00e9diction de la production v\u00e9g\u00e9tale<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Dans le domaine de la pr\u00e9vision de la production v\u00e9g\u00e9tale, les mod\u00e8les d&#039;IA ont exploit\u00e9 des donn\u00e9es environnementales telles que les conditions m\u00e9t\u00e9orologiques et les param\u00e8tres du sol pour pr\u00e9voir les rendements des cultures. Des mod\u00e8les de r\u00e9gression comme Isolation Forest, la r\u00e9gression lin\u00e9aire et la r\u00e9gression Ridge ont \u00e9t\u00e9 utilis\u00e9s.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Ces mod\u00e8les int\u00e9graient des caract\u00e9ristiques num\u00e9riques extraites des r\u00e9gions englobantes ainsi que des donn\u00e9es sur le sol afin d&#039;optimiser l&#039;application d&#039;engrais.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" width=\"810\" height=\"1076\" data-attachment-id=\"10632\" data-permalink=\"https:\/\/geopard.tech\/fr\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/sugar-beet-on-test-field\/\" data-orig-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-on-test-field.png?fit=1542%2C2048&amp;ssl=1\" data-orig-size=\"1542,2048\" data-comments-opened=\"1\" data-image-meta=\"{&quot;aperture&quot;:&quot;0&quot;,&quot;credit&quot;:&quot;&quot;,&quot;camera&quot;:&quot;&quot;,&quot;caption&quot;:&quot;&quot;,&quot;created_timestamp&quot;:&quot;0&quot;,&quot;copyright&quot;:&quot;&quot;,&quot;focal_length&quot;:&quot;0&quot;,&quot;iso&quot;:&quot;0&quot;,&quot;shutter_speed&quot;:&quot;0&quot;,&quot;title&quot;:&quot;&quot;,&quot;orientation&quot;:&quot;0&quot;}\" data-image-title=\"Sugar beet on test field\" data-image-description=\"\" data-image-caption=\"\" data-large-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-on-test-field.png?fit=771%2C1024&amp;ssl=1\" src=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-on-test-field.png?resize=810%2C1076&#038;ssl=1\" alt=\"Betterave sucri\u00e8re en champ d&#039;essai\" class=\"wp-image-10632\" style=\"width:700px;height:auto\" srcset=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-on-test-field.png?w=1542&amp;ssl=1 1542w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-on-test-field.png?resize=226%2C300&amp;ssl=1 226w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-on-test-field.png?resize=771%2C1024&amp;ssl=1 771w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-on-test-field.png?resize=768%2C1020&amp;ssl=1 768w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-on-test-field.png?resize=1157%2C1536&amp;ssl=1 1157w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-on-test-field.png?resize=150%2C200&amp;ssl=1 150w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Sugar-beet-on-test-field.png?resize=1200%2C1594&amp;ssl=1 1200w\" sizes=\"(max-width: 810px) 100vw, 810px\" \/><figcaption class=\"wp-element-caption\">Betterave sucri\u00e8re en champ d&#039;essai<\/figcaption><\/figure>\n<\/div>\n\n\n<h2 class=\"wp-block-heading\" id=\"h-model-deployment-considerations\">Consid\u00e9rations relatives au d\u00e9ploiement du mod\u00e8le<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Les strat\u00e9gies de d\u00e9ploiement des mod\u00e8les d\u00e9velopp\u00e9s ont \u00e9t\u00e9 \u00e9valu\u00e9es pour les dispositifs p\u00e9riph\u00e9riques et les plateformes cloud. Le d\u00e9ploiement des mod\u00e8les sur des dispositifs p\u00e9riph\u00e9riques a offert des avantages tels que des co\u00fbts r\u00e9duits et une latence plus faible.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Toutefois, cette approche peut nuire \u00e0 la pr\u00e9cision potentielle en raison des limitations mat\u00e9rielles. Par ailleurs, le d\u00e9ploiement dans le cloud offre des temps d&#039;inf\u00e9rence plus rapides gr\u00e2ce \u00e0 l&#039;utilisation de GPU haute performance, mais peut engendrer des co\u00fbts suppl\u00e9mentaires et d\u00e9pend d&#039;une connexion Internet, ce qui peut introduire une latence de communication.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"h-comparative-analysis-with-5g-network\">Analyse comparative avec le r\u00e9seau 5G<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Une analyse comparative a d\u00e9montr\u00e9 que l&#039;utilisation d&#039;un r\u00e9seau 5G am\u00e9liorait consid\u00e9rablement la segmentation des betteraves sucri\u00e8res par rapport aux configurations 4G\/WiFi traditionnelles. Cette am\u00e9lioration s&#039;est traduite par une r\u00e9duction des temps moyens de configuration et de connexion au r\u00e9seau, soulignant ainsi les gains d&#039;efficacit\u00e9 obtenus gr\u00e2ce \u00e0 la technologie 5G.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Processus de pr\u00e9paration des donn\u00e9es<\/strong><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Le processus de pr\u00e9paration des donn\u00e9es a consist\u00e9 \u00e0 collecter des ensembles de donn\u00e9es de plantes saines et malades, \u00e0 d\u00e9tecter les adventices, \u00e0 identifier les stades de croissance et \u00e0 extraire des images \u00e0 partir de vid\u00e9os brutes 4K. Des techniques telles que l&#039;\u00e9galisation d&#039;histogramme, le filtrage d&#039;images et la transformation de l&#039;espace colorim\u00e9trique HSV ont \u00e9t\u00e9 utilis\u00e9es pour pr\u00e9parer les donn\u00e9es \u00e0 l&#039;analyse.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Des \u00e9chantillons de feuilles de betterave sucri\u00e8re saines et de feuilles malades, comme des feuilles de ma\u00efs atteintes de cercosporiose, ont \u00e9t\u00e9 pr\u00e9lev\u00e9s. L&#039;extraction des caract\u00e9ristiques de la maladie a consist\u00e9 \u00e0 s\u00e9parer la feuille de son fond, \u00e0 redimensionner, transformer et fusionner les images afin de cr\u00e9er des \u00e9chantillons r\u00e9alistes pour l&#039;analyse.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" width=\"810\" height=\"200\" data-attachment-id=\"10633\" data-permalink=\"https:\/\/geopard.tech\/fr\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/annotation-process-for-segmentation\/\" data-orig-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Annotation-process-for-segmentation.png?fit=2048%2C505&amp;ssl=1\" data-orig-size=\"2048,505\" data-comments-opened=\"1\" data-image-meta=\"{&quot;aperture&quot;:&quot;0&quot;,&quot;credit&quot;:&quot;&quot;,&quot;camera&quot;:&quot;&quot;,&quot;caption&quot;:&quot;&quot;,&quot;created_timestamp&quot;:&quot;0&quot;,&quot;copyright&quot;:&quot;&quot;,&quot;focal_length&quot;:&quot;0&quot;,&quot;iso&quot;:&quot;0&quot;,&quot;shutter_speed&quot;:&quot;0&quot;,&quot;title&quot;:&quot;&quot;,&quot;orientation&quot;:&quot;0&quot;}\" data-image-title=\"Annotation process for segmentation\" data-image-description=\"\" data-image-caption=\"&lt;p&gt;Caption 3&lt;\/p&gt;\" data-large-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Annotation-process-for-segmentation.png?fit=1024%2C253&amp;ssl=1\" src=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Annotation-process-for-segmentation.png?resize=810%2C200&#038;ssl=1\" alt=\"Processus d&#039;annotation pour la segmentation\" class=\"wp-image-10633\" srcset=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Annotation-process-for-segmentation.png?w=2048&amp;ssl=1 2048w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Annotation-process-for-segmentation.png?resize=300%2C74&amp;ssl=1 300w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Annotation-process-for-segmentation.png?resize=1024%2C253&amp;ssl=1 1024w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Annotation-process-for-segmentation.png?resize=768%2C189&amp;ssl=1 768w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Annotation-process-for-segmentation.png?resize=1536%2C379&amp;ssl=1 1536w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Annotation-process-for-segmentation.png?resize=1200%2C296&amp;ssl=1 1200w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Annotation-process-for-segmentation.png?w=1620&amp;ssl=1 1620w\" sizes=\"(max-width: 810px) 100vw, 810px\" \/><figcaption class=\"wp-element-caption\">Processus d&#039;annotation pour la segmentation<\/figcaption><\/figure>\n<\/div>\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Boucle d&#039;apprentissage active<\/strong><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Un cycle d&#039;apprentissage actif a \u00e9t\u00e9 initi\u00e9 avec des donn\u00e9es non \u00e9tiquet\u00e9es, utilis\u00e9es pour entra\u00eener des mod\u00e8les de d\u00e9tection. Ces mod\u00e8les ont g\u00e9n\u00e9r\u00e9 des requ\u00eates d&#039;annotation auxquelles ont r\u00e9pondu des annotateurs humains, affinant ainsi continuellement la pr\u00e9cision du mod\u00e8le gr\u00e2ce \u00e0 des cycles it\u00e9ratifs d&#039;entra\u00eenement et d&#039;annotation.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Annotation des donn\u00e9es via un mod\u00e8le de base multimodal<\/strong><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Pour pallier le manque de donn\u00e9es \u00e9tiquet\u00e9es, le projet a exploit\u00e9 des mod\u00e8les de base robustes afin de g\u00e9n\u00e9rer des annotations de r\u00e9f\u00e9rence. En particulier, CLIP, un mod\u00e8le bas\u00e9 sur l&#039;architecture Transformer d\u00e9velopp\u00e9 par OpenAI et entra\u00een\u00e9 sur un vaste ensemble de donn\u00e9es de plus de 400 millions de paires image-texte, a jou\u00e9 un r\u00f4le d\u00e9terminant.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Utilisant Vision Transformers comme base, CLIP a atteint une pr\u00e9cision remarquable de 95% sur les ensembles de validation, cat\u00e9gorisant efficacement les images en classes distinctes telles que la betterave sucri\u00e8re et les mauvaises herbes avec une grande pr\u00e9cision.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Technologie des drones pour la collecte de donn\u00e9es<\/strong><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">L&#039;une des technologies cl\u00e9s utilis\u00e9es dans ce projet \u00e9tait le recours \u00e0 des drones \u00e9quip\u00e9s de cam\u00e9ras RVB qui capturaient des vid\u00e9os en 4K. Ces drones ont fourni des images d\u00e9taill\u00e9es (r\u00e9solution 3840\u00d72160) pour l&#039;analyse.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Le pr\u00e9traitement de ces images a consid\u00e9rablement am\u00e9lior\u00e9 la pr\u00e9cision du mod\u00e8le, avec des am\u00e9liorations notables observ\u00e9es dans des mod\u00e8les comme VGGNet (+38,52%), ResNet50 (+21,14%), DenseNet121 (+7,53%) et MobileNet (+6,6%).<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Des techniques telles que l&#039;\u00e9galisation d&#039;histogramme ont \u00e9t\u00e9 utilis\u00e9es pour am\u00e9liorer le contraste de l&#039;image, tandis que la transformation en espace colorim\u00e9trique HSV a permis de mettre en \u00e9vidence les zones v\u00e9g\u00e9tales et les caract\u00e9ristiques pertinentes.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>G\u00e9n\u00e9ration de donn\u00e9es synth\u00e9tiques<\/strong><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Pour pallier le manque de donn\u00e9es d&#039;images, des jeux de donn\u00e9es synth\u00e9tiques ont \u00e9t\u00e9 g\u00e9n\u00e9r\u00e9s gr\u00e2ce \u00e0 l&#039;apprentissage automatique et \u00e0 l&#039;intelligence artificielle. La collecte des donn\u00e9es a \u00e9t\u00e9 r\u00e9alis\u00e9e \u00e0 l&#039;aide de drones volant \u00e0 des hauteurs comprises entre 1 et 4 m\u00e8tres et \u00e0 des vitesses sup\u00e9rieures ou \u00e9gales \u00e0 2 m\/s, \u00e9quip\u00e9s de cam\u00e9ras RVB.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" width=\"810\" height=\"610\" data-attachment-id=\"10634\" data-permalink=\"https:\/\/geopard.tech\/fr\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/emulation-environment\/\" data-orig-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Emulation-environment.png?fit=2048%2C1542&amp;ssl=1\" data-orig-size=\"2048,1542\" data-comments-opened=\"1\" data-image-meta=\"{&quot;aperture&quot;:&quot;0&quot;,&quot;credit&quot;:&quot;&quot;,&quot;camera&quot;:&quot;&quot;,&quot;caption&quot;:&quot;&quot;,&quot;created_timestamp&quot;:&quot;0&quot;,&quot;copyright&quot;:&quot;&quot;,&quot;focal_length&quot;:&quot;0&quot;,&quot;iso&quot;:&quot;0&quot;,&quot;shutter_speed&quot;:&quot;0&quot;,&quot;title&quot;:&quot;&quot;,&quot;orientation&quot;:&quot;0&quot;}\" data-image-title=\"Emulation environment\" data-image-description=\"\" data-image-caption=\"\" data-large-file=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Emulation-environment.png?fit=1024%2C771&amp;ssl=1\" src=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Emulation-environment.png?resize=810%2C610&#038;ssl=1\" alt=\"Environnement d&#039;\u00e9mulation\" class=\"wp-image-10634\" style=\"width:646px;height:auto\" srcset=\"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Emulation-environment.png?w=2048&amp;ssl=1 2048w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Emulation-environment.png?resize=300%2C226&amp;ssl=1 300w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Emulation-environment.png?resize=1024%2C771&amp;ssl=1 1024w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Emulation-environment.png?resize=768%2C578&amp;ssl=1 768w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Emulation-environment.png?resize=1536%2C1157&amp;ssl=1 1536w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Emulation-environment.png?resize=400%2C300&amp;ssl=1 400w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Emulation-environment.png?resize=200%2C150&amp;ssl=1 200w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Emulation-environment.png?resize=1200%2C904&amp;ssl=1 1200w, https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/Emulation-environment.png?w=1620&amp;ssl=1 1620w\" sizes=\"(max-width: 810px) 100vw, 810px\" \/><figcaption class=\"wp-element-caption\">Environnement d&#039;\u00e9mulation<\/figcaption><\/figure>\n<\/div>\n\n\n<p class=\"wp-block-paragraph\">D&#039;autres v\u00e9hicules, comme des tracteurs, ont \u00e9galement \u00e9t\u00e9 utilis\u00e9s pour la collecte de donn\u00e9es. Cette g\u00e9n\u00e9ration de donn\u00e9es synth\u00e9tiques s&#039;est av\u00e9r\u00e9e particuli\u00e8rement utile pour la d\u00e9tection des maladies de la betterave sucri\u00e8re.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"h-conclusion\">Conclusion<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Le projet \u201c\u00a0R\u00e9seaux 5G\u00a0: un catalyseur d\u2019apprentissage en temps r\u00e9el pour une agriculture durable\u00a0\u201d a d\u00e9montr\u00e9 avec succ\u00e8s comment la technologie 5G peut am\u00e9liorer les aspects \u00e9cologiques, \u00e9conomiques et durables de la culture de la betterave sucri\u00e8re. Gr\u00e2ce \u00e0 une collaboration avec HSHL et Pfeifer &amp; Langen, le projet a int\u00e9gr\u00e9 la collecte de donn\u00e9es en temps r\u00e9el et l\u2019analyse pilot\u00e9e par l\u2019IA, ce qui a permis d\u2019accro\u00eetre l\u2019efficacit\u00e9 et de r\u00e9duire les d\u00e9placements inutiles sur le terrain. <\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Un r\u00e9seau 5G d\u00e9di\u00e9 sur le campus a permis une application pr\u00e9cise d&#039;engrais et de produits phytosanitaires. Geopard Agriculture a jou\u00e9 un r\u00f4le crucial dans le d\u00e9veloppement de sc\u00e9narios de d\u00e9tection et de surveillance des plantes, ainsi que dans la cr\u00e9ation d&#039;un prototype de syst\u00e8me d&#039;apprentissage automatique pour l&#039;environnement agricole 5G. Le succ\u00e8s de ce projet a soulign\u00e9 l&#039;importance des technologies de pointe pour une agriculture durable, mettant en lumi\u00e8re le potentiel de la 5G pour stimuler l&#039;innovation et l&#039;efficacit\u00e9. <\/p>","protected":false},"excerpt":{"rendered":"<p>Nous sommes ravis d&#039;annoncer la r\u00e9ussite du projet \u201c\u00a0Les r\u00e9seaux 5G comme catalyseur d&#039;apprentissage en temps r\u00e9el dans l&#039;agriculture durable\u00a0\u201d, soutenu en partie par\u2026<\/p>","protected":false},"author":210249433,"featured_media":10639,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_coblocks_attr":"","_coblocks_dimensions":"","_coblocks_responsive_height":"","_coblocks_accordion_ie_support":"","_eb_attr":"","_crdt_document":"","content-type":"","jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","enabled":false},"version":2},"_wpas_customize_per_network":false},"categories":[1657,1367],"tags":[],"class_list":["post-10617","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-precision-farming","category-use-cases"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v21.6 (Yoast SEO v27.3) - https:\/\/yoast.com\/product\/yoast-seo-premium-wordpress\/ -->\n<title>5G-Enabled Real-time Learning in Sustainable Farming: A Study on Sugar Beet - GeoPard Agriculture<\/title>\n<meta name=\"description\" content=\"Successful completion of &quot;5G Networks as an Enabler for Real-time Learning in Sustainable Farming&quot; project, supported by State of North Rhine-Westphalia.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/geopard.tech\/fr\/blog\/la-5g-a-permis-un-apprentissage-en-temps-reel-dans-lagriculture-durable-une-etude-sur-la-production-de-betteraves-sucrieres\/\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"5G-Enabled Real-time Learning in Sustainable Farming: A Study on Sugar Beet\" \/>\n<meta property=\"og:description\" content=\"Successful completion of &quot;5G Networks as an Enabler for Real-time Learning in Sustainable Farming&quot; project, supported by State of North Rhine-Westphalia.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/geopard.tech\/fr\/blog\/la-5g-a-permis-un-apprentissage-en-temps-reel-dans-lagriculture-durable-une-etude-sur-la-production-de-betteraves-sucrieres\/\" \/>\n<meta property=\"og:site_name\" content=\"GeoPard - Precision agriculture Mapping software\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/geopardAgriculture\/\" \/>\n<meta property=\"article:published_time\" content=\"2024-07-13T18:50:45+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-07-13T18:50:51+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/geopard.tech\/wp-content\/uploads\/2024\/07\/5G-Enabled-Real-time-Learning-in-Sustainable-Farming-A-Study-on-Sugar-Beet-1024x576.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1024\" \/>\n\t<meta property=\"og:image:height\" content=\"576\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Muhammad Farjad\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@geopardagri\" \/>\n<meta name=\"twitter:site\" content=\"@geopardagri\" \/>\n<meta name=\"twitter:label1\" content=\"\u00c9crit par\" \/>\n\t<meta name=\"twitter:data1\" content=\"Muhammad Farjad\" \/>\n\t<meta name=\"twitter:label2\" content=\"Dur\u00e9e de lecture estim\u00e9e\" \/>\n\t<meta name=\"twitter:data2\" content=\"11 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\\\/\"},\"author\":{\"name\":\"Muhammad Farjad\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/#\\\/schema\\\/person\\\/123c5562fb47aa8cf3aa81ae91e5e935\"},\"headline\":\"5G-Enabled Real-time Learning in Sustainable Farming: A Study on Sugar Beet\",\"datePublished\":\"2024-07-13T18:50:45+00:00\",\"dateModified\":\"2024-07-13T18:50:51+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\\\/\"},\"wordCount\":2143,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2024\\\/07\\\/5G-Enabled-Real-time-Learning-in-Sustainable-Farming-A-Study-on-Sugar-Beet.png?fit=1920%2C1080&ssl=1\",\"articleSection\":[\"Precision Farming\",\"Use-cases\"],\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/geopard.tech\\\/blog\\\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\\\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\\\/\",\"url\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\\\/\",\"name\":\"5G-Enabled Real-time Learning in Sustainable Farming: A Study on Sugar Beet - GeoPard Agriculture\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2024\\\/07\\\/5G-Enabled-Real-time-Learning-in-Sustainable-Farming-A-Study-on-Sugar-Beet.png?fit=1920%2C1080&ssl=1\",\"datePublished\":\"2024-07-13T18:50:45+00:00\",\"dateModified\":\"2024-07-13T18:50:51+00:00\",\"description\":\"Successful completion of \\\"5G Networks as an Enabler for Real-time Learning in Sustainable Farming\\\" project, supported by State of North Rhine-Westphalia.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\\\/#breadcrumb\"},\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/geopard.tech\\\/blog\\\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\\\/#primaryimage\",\"url\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2024\\\/07\\\/5G-Enabled-Real-time-Learning-in-Sustainable-Farming-A-Study-on-Sugar-Beet.png?fit=1920%2C1080&ssl=1\",\"contentUrl\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2024\\\/07\\\/5G-Enabled-Real-time-Learning-in-Sustainable-Farming-A-Study-on-Sugar-Beet.png?fit=1920%2C1080&ssl=1\",\"width\":1920,\"height\":1080,\"caption\":\"Caption 2\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/geopard.tech\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"5G-Enabled Real-time Learning in Sustainable Farming: A Study on Sugar Beet\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/#website\",\"url\":\"https:\\\/\\\/geopard.tech\\\/\",\"name\":\"GeoPard - Precision agriculture software\",\"description\":\"Precision agriculture Mapping software\",\"publisher\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/#organization\"},\"alternateName\":\"GeoPard\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/geopard.tech\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"fr-FR\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/#organization\",\"name\":\"GeoPard Agriculture\",\"alternateName\":\"GeoPard\",\"url\":\"https:\\\/\\\/geopard.tech\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2022\\\/03\\\/favicon.png?fit=200%2C200&ssl=1\",\"contentUrl\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2022\\\/03\\\/favicon.png?fit=200%2C200&ssl=1\",\"width\":200,\"height\":200,\"caption\":\"GeoPard Agriculture\"},\"image\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/www.facebook.com\\\/geopardAgriculture\\\/\",\"https:\\\/\\\/x.com\\\/geopardagri\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/geopard-agriculture\\\/\",\"https:\\\/\\\/www.instagram.com\\\/geopardagriculture\\\/\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/#\\\/schema\\\/person\\\/123c5562fb47aa8cf3aa81ae91e5e935\",\"name\":\"Muhammad Farjad\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/51b9af87d9b0801ae1c0ab294a0f06eb669f60d5168f33394aeeca9de86537bb?s=96&d=identicon&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/51b9af87d9b0801ae1c0ab294a0f06eb669f60d5168f33394aeeca9de86537bb?s=96&d=identicon&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/51b9af87d9b0801ae1c0ab294a0f06eb669f60d5168f33394aeeca9de86537bb?s=96&d=identicon&r=g\",\"caption\":\"Muhammad Farjad\"},\"url\":\"#\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Apprentissage en temps r\u00e9el gr\u00e2ce \u00e0 la 5G pour une agriculture durable\u00a0: une \u00e9tude sur la betterave sucri\u00e8re - GeoPard Agriculture","description":"R\u00e9alisation r\u00e9ussie du projet \" Les r\u00e9seaux 5G comme catalyseur d&#039;apprentissage en temps r\u00e9el dans l&#039;agriculture durable \", soutenu par le Land de Rh\u00e9nanie-du-Nord-Westphalie.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/geopard.tech\/fr\/blog\/la-5g-a-permis-un-apprentissage-en-temps-reel-dans-lagriculture-durable-une-etude-sur-la-production-de-betteraves-sucrieres\/","og_locale":"fr_FR","og_type":"article","og_title":"5G-Enabled Real-time Learning in Sustainable Farming: A Study on Sugar Beet","og_description":"Successful completion of \"5G Networks as an Enabler for Real-time Learning in Sustainable Farming\" project, supported by State of North Rhine-Westphalia.","og_url":"https:\/\/geopard.tech\/fr\/blog\/la-5g-a-permis-un-apprentissage-en-temps-reel-dans-lagriculture-durable-une-etude-sur-la-production-de-betteraves-sucrieres\/","og_site_name":"GeoPard - Precision agriculture Mapping software","article_publisher":"https:\/\/www.facebook.com\/geopardAgriculture\/","article_published_time":"2024-07-13T18:50:45+00:00","article_modified_time":"2024-07-13T18:50:51+00:00","og_image":[{"width":1024,"height":576,"url":"https:\/\/geopard.tech\/wp-content\/uploads\/2024\/07\/5G-Enabled-Real-time-Learning-in-Sustainable-Farming-A-Study-on-Sugar-Beet-1024x576.png","type":"image\/png"}],"author":"Muhammad Farjad","twitter_card":"summary_large_image","twitter_creator":"@geopardagri","twitter_site":"@geopardagri","twitter_misc":{"\u00c9crit par":"Muhammad Farjad","Dur\u00e9e de lecture estim\u00e9e":"11 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/geopard.tech\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/#article","isPartOf":{"@id":"https:\/\/geopard.tech\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/"},"author":{"name":"Muhammad Farjad","@id":"https:\/\/geopard.tech\/#\/schema\/person\/123c5562fb47aa8cf3aa81ae91e5e935"},"headline":"5G-Enabled Real-time Learning in Sustainable Farming: A Study on Sugar Beet","datePublished":"2024-07-13T18:50:45+00:00","dateModified":"2024-07-13T18:50:51+00:00","mainEntityOfPage":{"@id":"https:\/\/geopard.tech\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/"},"wordCount":2143,"commentCount":0,"publisher":{"@id":"https:\/\/geopard.tech\/#organization"},"image":{"@id":"https:\/\/geopard.tech\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/#primaryimage"},"thumbnailUrl":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/5G-Enabled-Real-time-Learning-in-Sustainable-Farming-A-Study-on-Sugar-Beet.png?fit=1920%2C1080&ssl=1","articleSection":["Precision Farming","Use-cases"],"inLanguage":"fr-FR","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/geopard.tech\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/geopard.tech\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/","url":"https:\/\/geopard.tech\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/","name":"Apprentissage en temps r\u00e9el gr\u00e2ce \u00e0 la 5G pour une agriculture durable\u00a0: une \u00e9tude sur la betterave sucri\u00e8re - GeoPard Agriculture","isPartOf":{"@id":"https:\/\/geopard.tech\/#website"},"primaryImageOfPage":{"@id":"https:\/\/geopard.tech\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/#primaryimage"},"image":{"@id":"https:\/\/geopard.tech\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/#primaryimage"},"thumbnailUrl":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/5G-Enabled-Real-time-Learning-in-Sustainable-Farming-A-Study-on-Sugar-Beet.png?fit=1920%2C1080&ssl=1","datePublished":"2024-07-13T18:50:45+00:00","dateModified":"2024-07-13T18:50:51+00:00","description":"R\u00e9alisation r\u00e9ussie du projet \" Les r\u00e9seaux 5G comme catalyseur d&#039;apprentissage en temps r\u00e9el dans l&#039;agriculture durable \", soutenu par le Land de Rh\u00e9nanie-du-Nord-Westphalie.","breadcrumb":{"@id":"https:\/\/geopard.tech\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/#breadcrumb"},"inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/geopard.tech\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/"]}]},{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/geopard.tech\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/#primaryimage","url":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/5G-Enabled-Real-time-Learning-in-Sustainable-Farming-A-Study-on-Sugar-Beet.png?fit=1920%2C1080&ssl=1","contentUrl":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/5G-Enabled-Real-time-Learning-in-Sustainable-Farming-A-Study-on-Sugar-Beet.png?fit=1920%2C1080&ssl=1","width":1920,"height":1080,"caption":"Caption 2"},{"@type":"BreadcrumbList","@id":"https:\/\/geopard.tech\/blog\/5g-enabled-real-time-learning-in-sustainable-farming-a-study-on-sugar-beet-production\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/geopard.tech\/"},{"@type":"ListItem","position":2,"name":"5G-Enabled Real-time Learning in Sustainable Farming: A Study on Sugar Beet"}]},{"@type":"WebSite","@id":"https:\/\/geopard.tech\/#website","url":"https:\/\/geopard.tech\/","name":"GeoPard - Logiciel d'agriculture de pr\u00e9cision","description":"Logiciel de cartographie pour l'agriculture de pr\u00e9cision","publisher":{"@id":"https:\/\/geopard.tech\/#organization"},"alternateName":"GeoPard","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/geopard.tech\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"fr-FR"},{"@type":"Organization","@id":"https:\/\/geopard.tech\/#organization","name":"GeoPard Agriculture","alternateName":"GeoPard","url":"https:\/\/geopard.tech\/","logo":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/geopard.tech\/#\/schema\/logo\/image\/","url":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2022\/03\/favicon.png?fit=200%2C200&ssl=1","contentUrl":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2022\/03\/favicon.png?fit=200%2C200&ssl=1","width":200,"height":200,"caption":"GeoPard Agriculture"},"image":{"@id":"https:\/\/geopard.tech\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/geopardAgriculture\/","https:\/\/x.com\/geopardagri","https:\/\/www.linkedin.com\/company\/geopard-agriculture\/","https:\/\/www.instagram.com\/geopardagriculture\/"]},{"@type":"Person","@id":"https:\/\/geopard.tech\/#\/schema\/person\/123c5562fb47aa8cf3aa81ae91e5e935","name":"Muhammad Farjad","image":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/secure.gravatar.com\/avatar\/51b9af87d9b0801ae1c0ab294a0f06eb669f60d5168f33394aeeca9de86537bb?s=96&d=identicon&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/51b9af87d9b0801ae1c0ab294a0f06eb669f60d5168f33394aeeca9de86537bb?s=96&d=identicon&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/51b9af87d9b0801ae1c0ab294a0f06eb669f60d5168f33394aeeca9de86537bb?s=96&d=identicon&r=g","caption":"Muhammad Farjad"},"url":"#"}]}},"jetpack_publicize_connections":[],"jetpack_featured_media_url":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2024\/07\/5G-Enabled-Real-time-Learning-in-Sustainable-Farming-A-Study-on-Sugar-Beet.png?fit=1920%2C1080&ssl=1","jetpack_likes_enabled":true,"jetpack_sharing_enabled":true,"jetpack_shortlink":"https:\/\/wp.me\/pdiCPa-2Lf","_links":{"self":[{"href":"https:\/\/geopard.tech\/fr\/wp-json\/wp\/v2\/posts\/10617","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/geopard.tech\/fr\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/geopard.tech\/fr\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/geopard.tech\/fr\/wp-json\/wp\/v2\/users\/210249433"}],"replies":[{"embeddable":true,"href":"https:\/\/geopard.tech\/fr\/wp-json\/wp\/v2\/comments?post=10617"}],"version-history":[{"count":0,"href":"https:\/\/geopard.tech\/fr\/wp-json\/wp\/v2\/posts\/10617\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/geopard.tech\/fr\/wp-json\/wp\/v2\/media\/10639"}],"wp:attachment":[{"href":"https:\/\/geopard.tech\/fr\/wp-json\/wp\/v2\/media?parent=10617"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/geopard.tech\/fr\/wp-json\/wp\/v2\/categories?post=10617"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/geopard.tech\/fr\/wp-json\/wp\/v2\/tags?post=10617"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}