{"id":11559,"date":"2025-05-04T22:58:53","date_gmt":"2025-05-04T20:58:53","guid":{"rendered":"https:\/\/geopard.tech\/?p=11559"},"modified":"2025-05-04T23:00:29","modified_gmt":"2025-05-04T21:00:29","slug":"gojenje-jecmena-dobi-zagon-z-lahkim-zaznavanjem-yolov5","status":"publish","type":"post","link":"https:\/\/geopard.tech\/sl\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/","title":{"rendered":"Gojenje je\u010dmena pridobiva zagon z lahkoto zaznavanja YOLOv5"},"content":{"rendered":"

\u0160kotski je\u010dmen, odporna \u017eitna kultura, ki jo gojijo v visokogorskih obmo\u010djih kitajske planote Qinghai-Tibet, igra klju\u010dno vlogo pri lokalni prehranski varnosti in gospodarski stabilnosti. Znanstveno znan kot\u00a0Hordeum vulgare<\/em>\u00a0L., ta pridelek uspeva v ekstremnih razmerah \u2013 redkem zraku, nizki ravni kisika in povpre\u010dni letni temperaturi 6,3 \u00b0C \u2013 zaradi \u010desar je nepogre\u0161ljiv za skupnosti v te\u017ekih okoljih.<\/p>\n

Z ve\u010d kot 270.000 hektarji, namenjenimi gojenju na Kitajskem, predvsem v avtonomni regiji Xizang, predstavlja visokogorski je\u010dmen ve\u010d kot polovico posejanih povr\u0161in v regiji in ve\u010d kot 70% celotne proizvodnje \u017eita. Natan\u010dno spremljanje gostote je\u010dmena \u2013 \u0161tevila rastlin ali klasov na enoto povr\u0161ine \u2013 je bistvenega pomena za optimizacijo kmetijskih praks, kot sta namakanje in gnojenje, ter napovedovanje pridelkov.<\/p>\n

Vendar so se tradicionalne metode, kot sta ro\u010dno vzor\u010denje ali satelitsko slikanje, izkazale za neu\u010dinkovite, delovno intenzivne ali premalo podrobne. Za re\u0161evanje teh izzivov so raziskovalci z Univerze za kmetijstvo in gozdarstvo Fujian in Tehnolo\u0161ke univerze Chengdu razvili inovativen model umetne inteligence, ki temelji na YOLOv5, vrhunskem algoritmu za zaznavanje objektov.<\/p>\n

Njihovo delo, objavljeno v\u00a0Metode rastlin<\/em>\u00a0(2025) je dosegel izjemne rezultate, vklju\u010dno s povpre\u010dno natan\u010dnostjo (mAP) 93,1% \u2013 metriko, ki meri splo\u0161no natan\u010dnost zaznavanja \u2013 in zmanj\u0161anjem ra\u010dunskih stro\u0161kov za 75,6%, zaradi \u010desar je primeren za uporabo dronov v realnem \u010dasu.<\/p>\n

Izzivi in inovacije pri spremljanju pridelkov<\/h2>\n

Pomen visokogorskega je\u010dmena sega dlje od njegove vloge kot vira hrane. Samo leta 2022 je mesto Rikaze, pomembna regija za pridelavo je\u010dmena, po\u017eelo 408.900 ton je\u010dmena na 60.000 hektarjih, kar je skoraj polovica celotne proizvodnje \u017eita v Tibetu.<\/p>\n

Kljub kulturnemu in gospodarskemu pomenu je ocenjevanje pridelka je\u010dmena \u017ee dolgo izziv. Tradicionalne metode, kot sta ro\u010dno \u0161tetje ali satelitski posnetki, so bodisi preve\u010d delovno intenzivne bodisi nimajo dovolj lo\u010dljivosti, da bi zaznali posamezne klasje je\u010dmena \u2013 del rastline, ki nosi zrnje in je pogosto \u0161irok le 2\u20133 centimetre.<\/p>\n

Ro\u010dno vzor\u010denje od kmetov zahteva fizi\u010dni pregled delov polja \u2013 postopek, ki je po\u010dasen, subjektiven in neprakti\u010den za velike kmetije. Satelitski posnetki so sicer uporabni za obse\u017ena opazovanja, vendar se soo\u010dajo z nizko lo\u010dljivostjo (pogosto 10\u201330 metrov na slikovno piko) in pogostimi vremenskimi motnjami, kot je obla\u010dnost v gorskih regijah, kot je Tibet.<\/p>\n

Da bi premagali te omejitve, so se raziskovalci obrnili na brezpilotna letala (UAV) oziroma drone, opremljene z 20-megapikselnimi kamerami. Ti droni so posneli 501 visokolo\u010dljivostno sliko je\u010dmenovih polj v mestu Rikaze med dvema kriti\u010dnima fazama rasti: fazo rasti avgusta 2022, za katero so zna\u010dilni zeleni, razvijajo\u010di se trni, in fazo zorenja avgusta 2023, za katero so zna\u010dilni zlato rumeni trni, pripravljeni za \u017eetev.<\/p>\n

\"Spremljanje<\/p>\n

Vendar pa je analiza teh slik predstavljala izzive, vklju\u010dno z zamegljenimi robovi, ki jih povzro\u010da gibanje drona, majhnostjo je\u010dmenovih klasov na posnetkih iz zraka in prekrivajo\u010dimi se klasmi na gosto zasajenih poljih.<\/p>\n

Da bi re\u0161ili te te\u017eave, so raziskovalci slike predhodno obdelali tako, da so vsako sliko visoke lo\u010dljivosti razdelili na 35 manj\u0161ih podslik in filtrirali zamegljene robove, kar je privedlo do 2970 visokokakovostnih podslik za u\u010denje. Ta korak predhodne obdelave je zagotovil, da se je model osredoto\u010dil na jasne in uporabne podatke, s \u010dimer se je izognil motnjam zaradi obmo\u010dij nizke kakovosti.<\/p>\n

Tehni\u010dni napredek pri zaznavanju objektov<\/h2>\n

Osrednji del te raziskave je algoritem YOLOv5 (You Only Look Once version 5), enostopenjski model zaznavanja objektov, znan po svoji hitrosti in modularni zasnovi. Za razliko od starej\u0161ih dvostopenjskih modelov, kot je Faster R-CNN, ki najprej identificirajo obmo\u010dja zanimanja in nato razvrstijo objekte, YOLOv5 izvede zaznavanje v enem samem prehodu, zaradi \u010desar je bistveno hitrej\u0161i.<\/p>\n

Osnovni model YOLOv5n z 1,76 milijona parametrov (nastavljive komponente modela umetne inteligence) in 4,1 milijarde FLOP-ov (operacije s plavajo\u010do vejico, mera ra\u010dunske kompleksnosti) je bil \u017ee u\u010dinkovit. Vendar pa je zaznavanje drobnih, prekrivajo\u010dih se konic je\u010dmena zahtevalo nadaljnjo optimizacijo.<\/p>\n

Raziskovalna ekipa je modelu predstavila tri klju\u010dne izbolj\u0161ave: globinsko lo\u010dljivo konvolucijo (DSConv), duhovno konvolucijo (GhostConv) in modul za pozornost konvolucijskih blokov (CBAM).<\/p>\n

Globinsko lo\u010dljiva konvolucija (DSConv) zmanj\u0161a ra\u010dunske stro\u0161ke z razdelitvijo standardnega procesa konvolucije \u2013 matemati\u010dne operacije, ki iz slik izlu\u0161\u010di zna\u010dilnosti \u2013 na dva koraka. Najprej globinska konvolucija uporabi filtre za posamezne barvne kanale (npr. rde\u010do, zeleno, modro) in analizira vsak kanal posebej.<\/p>\n

Sledi to\u010dkovna konvolucija, ki zdru\u017euje rezultate po kanalih z uporabo jeder 1\u00d71. Ta pristop zmanj\u0161a \u0161tevilo parametrov za do 75%.<\/p>\n

\"Zmanj\u0161anje<\/p>\n

Na primer, tradicionalna konvolucija 3\u00d73 s 64 vhodnimi in 128 izhodnimi kanali zahteva 73.728 parametrov, medtem ko DSConv to zmanj\u0161a na le 8.768 \u2013 zmanj\u0161anje za 88%. Ta u\u010dinkovitost je klju\u010dnega pomena za uporabo modelov na dronih ali mobilnih napravah z omejeno procesorsko mo\u010djo.<\/p>\n

Ghostna konvolucija (GhostConv) dodatno olaj\u0161a model z ustvarjanjem dodatnih zemljevidov zna\u010dilnosti \u2013 poenostavljenih predstavitev vzorcev slik \u2013 z uporabo preprostih linearnih operacij, kot sta rotacija ali skaliranje, namesto konvolucij, ki zahtevajo veliko virov.<\/p>\n

Tradicionalne konvolucijske plasti ustvarjajo odve\u010dne funkcije, kar porablja ra\u010dunalni\u0161ke vire. GhostConv to re\u0161uje tako, da iz obstoje\u010dih funkcij ustvari \u201cfantomske\u201d funkcije in s tem u\u010dinkovito prepolovi parametre v dolo\u010denih plasteh.<\/p>\n

Na primer, plast s 64 vhodnimi in 128 izhodnimi kanali bi tradicionalno zahtevala\u00a073.728 parametrov<\/strong>, vendar GhostConv to zmanj\u0161a na\u00a036,864<\/strong>\u00a0hkrati pa ohranja natan\u010dnost. Ta tehnika je \u0161e posebej uporabna za zaznavanje majhnih predmetov, kot so je\u010dmenovi klasji, kjer je ra\u010dunska u\u010dinkovitost izjemnega pomena.<\/p>\n

Modul za konvolucijsko blokovno pozornost (CBAM) je bil integriran, da bi modelu pomagal osredoto\u010diti se na kriti\u010dne zna\u010dilnosti, tudi v natrpanih okoljih. Mehanizmi pozornosti, ki jih navdihujejo \u010dlove\u0161ki vidni sistemi, modelom umetne inteligence omogo\u010dajo, da prednostno obravnavajo pomembne dele slike.<\/p>\n

CBAM uporablja dve vrsti pozornosti: kanalsko pozornost, ki prepozna pomembne barvne kanale (npr. zeleno za rasto\u010de klasje), in prostorsko pozornost, ki poudarja klju\u010dna obmo\u010dja znotraj slike (npr. skupine klasjev). Z zamenjavo standardnih modulov z DSConv in GhostConv ter vklju\u010ditvijo CBAM so raziskovalci ustvarili vitkej\u0161i in natan\u010dnej\u0161i model, prilagojen za zaznavanje je\u010dmena.<\/p>\n

Izvajanje in rezultati<\/h2>\n

Za u\u010denje modela so raziskovalci ro\u010dno ozna\u010dili 135 izvirnih slik z omejevalnimi okvirji \u2013 pravokotnimi okvirji, ki ozna\u010dujejo lokacijo klasov je\u010dmena \u2013 in kategorizirali klasje v faze rasti in zorenja. Tehnike dopolnjevanja podatkov \u2013 vklju\u010dno z rotacijo, vbrizgavanjem \u0161uma, okluzijo in ostrenjem \u2013 so raz\u0161irile nabor podatkov na 2970 slik, s \u010dimer so izbolj\u0161ale sposobnost modela za posplo\u0161evanje v razli\u010dnih terenskih pogojih.<\/p>\n

Na primer, vrtenje slik za 90\u00b0, 180\u00b0 ali 270\u00b0 je modelu pomagalo prepoznati konice iz razli\u010dnih kotov, hkrati pa je dodal \u0161um, ki je simuliral nepopolnosti iz resni\u010dnega sveta, kot sta prah ali sence. Nabor podatkov je bil razdeljen na u\u010dni nabor (80%) in validacijski nabor (20%), kar je zagotovilo robustno vrednotenje.<\/p>\n

Usposabljanje je potekalo na visokozmogljivem sistemu s procesorjem AMD Ryzen 7, grafi\u010dno kartico NVIDIA RTX 4060 in 64 GB RAM-a, z uporabo ogrodja PyTorch \u2013 priljubljenega orodja za globoko u\u010denje. Skrbno je bilo spremljanih ve\u010d kot 300 u\u010dnih epoh (popolnih prehodov skozi nabor podatkov), natan\u010dnost modela (natan\u010dnost pravilnih zaznav), priklic (sposobnost najti vse relevantne konice) in izguba (stopnja napak).<\/p>\n

Rezultati so bili osupljivi. Izbolj\u0161an model YOLOv5 je dosegel natan\u010dnost 92,2% (v primerjavi z 89,1% v izhodi\u0161\u010du) in odpoklic 86,2% (v primerjavi z 83,1%), s \u010dimer je v obeh metrikah presegel izhodi\u0161\u010dni model YOLOv5n za 3,1%. Njegova povpre\u010dna natan\u010dnost (mAP) \u2013 celovita metrika, ki povpre\u010di natan\u010dnost zaznavanja v vseh kategorijah \u2013 je dosegla 93,1%, s posameznimi rezultati 92,7% za skoke v fazi rasti in 93,5% za skoke v fazi zorenja.<\/p>\n

\"Rezultati<\/p>\n

Enako impresivna je bila njegova ra\u010dunska u\u010dinkovitost: parametri modela so se zmanj\u0161ali za 70,6% na 1,2 milijona, \u0161tevilo FLOP-ov pa za 75,6% na 3,1 milijarde. Primerjalne analize z vodilnimi modeli, kot sta Faster R-CNN in YOLOv8n, so poudarile njegovo superiornost.<\/p>\n

Medtem ko je YOLOv8n dosegel nekoliko vi\u0161ji mAP (93,8%), so bili njegovi parametri (3,0 milijona) in FLOP-i (8,1 milijarde) 2,5-krat oziroma 2,6-krat vi\u0161ji, zaradi \u010desar je predlagani model veliko u\u010dinkovitej\u0161i za aplikacije v realnem \u010dasu.<\/p>\n

Vizualne primerjave so poudarile ta napredek. Na slikah v fazi rasti je izbolj\u0161ani model zaznal 41 konic v primerjavi z 28 v osnovni fazi. Med zorenjem je prepoznal 3 konice v primerjavi z dvema v osnovni fazi, z manj zgre\u0161enimi zaznavami (ozna\u010denimi z oran\u017enimi pu\u0161\u010dicami) in la\u017eno pozitivnimi rezultati (ozna\u010denimi z vijoli\u010dnimi pu\u0161\u010dicami).<\/p>\n

Te izbolj\u0161ave so klju\u010dne za kmete, ki se zana\u0161ajo na natan\u010dne podatke za napovedovanje pridelkov in optimizacijo virov. Natan\u010dno \u0161tetje klasov na primer omogo\u010da bolj\u0161e ocene pridelave \u017eita, kar omogo\u010da sprejemanje odlo\u010ditev o \u010dasu \u017eetve, skladi\u0161\u010denju in na\u010drtovanju trga.<\/p>\n

Prihodnje smeri in prakti\u010dne posledice<\/h2>\n

Kljub uspehu je \u0161tudija priznala omejitve. Zmogljivost se je zmanj\u0161ala v ekstremnih svetlobnih pogojih, kot so ostro opoldansko ble\u0161\u010danje ali mo\u010dne sence, ki lahko zakrijejo podrobnosti konic. Poleg tega pravokotni omejevalni okvirji v\u010dasih niso ustrezali nepravilno oblikovanim konicam, kar je povzro\u010dilo manj\u0161e neto\u010dnosti.<\/p>\n

Model je iz slik brezpilotnih letalnikov izklju\u010dil tudi zamegljene robove, kar je zahtevalo ro\u010dno predobdelavo \u2013 korak, ki pove\u010duje \u010das in kompleksnost.<\/p>\n

Prihodnje delo si prizadeva re\u0161iti te te\u017eave z raz\u0161iritvijo nabora podatkov, da bi vklju\u010deval slike, posnete ob zori, opoldne in mraku, eksperimentiranjem z opombami v obliki poligonov (prilagodljive oblike, ki se bolje prilegajo nepravilnim predmetom) in razvojem algoritmov za bolj\u0161e obravnavo zamegljenih obmo\u010dij brez ro\u010dnega posredovanja.<\/p>\n

Posledice te raziskave so obse\u017ene. Za kmete v regijah, kot je Tibet, model ponuja oceno pridelka v realnem \u010dasu in nadome\u0161\u010da delovno intenzivno ro\u010dno \u0161tetje z avtomatizacijo, ki temelji na dronih. Razlikovanje med fazami rasti omogo\u010da natan\u010dno na\u010drtovanje \u017eetve in zmanj\u0161uje izgube zaradi prezgodnje ali zapoznele \u017eetve.<\/p>\n

Podrobni podatki o gostoti klasov \u2013 kot je prepoznavanje premalo poseljenih ali prenaseljenih obmo\u010dij \u2013 lahko pomagajo pri oblikovanju strategij namakanja in gnojenja, s \u010dimer se zmanj\u0161a poraba vode in kemikalij. Poleg je\u010dmena je lahka arhitektura obetavna tudi za druge polj\u0161\u010dine, kot so p\u0161enica, ri\u017e ali sadje, kar utira pot \u0161ir\u0161i uporabi v preciznem kmetijstvu.<\/p>\n

Zaklju\u010dek<\/h2>\n

Skratka, ta \u0161tudija ponazarja transformativni potencial umetne inteligence pri re\u0161evanju kmetijskih izzivov. Z izpopolnjevanjem YOLOv5 z inovativnimi lahkimi tehnikami so raziskovalci ustvarili orodje, ki uravnote\u017ei natan\u010dnost in u\u010dinkovitost \u2013 kar je klju\u010dnega pomena za uporabo v resni\u010dnem svetu v okoljih z omejenimi viri.<\/p>\n

Izrazi, kot so mAP, FLOP in mehanizmi pozornosti, se morda zdijo tehni\u010dni, vendar je njihov vpliv zelo prakti\u010den: kmetom omogo\u010dajo sprejemanje odlo\u010ditev na podlagi podatkov, ohranjanje virov in maksimiranje donosov. Ker podnebne spremembe in rast prebivalstva pove\u010dujejo pritisk na svetovne prehranske sisteme, bodo tak\u0161ni napredki nepogre\u0161ljivi.<\/p>\n

Za kmete v Tibetu in drugod ta tehnologija ne predstavlja le skoka v kmetijski u\u010dinkovitosti, temve\u010d tudi svetilnik upanja za trajnostno prehransko varnost v negotovi prihodnosti.<\/p>\n

Referenca: <\/strong>Cai, M., Deng, H., Cai, J. et al. Zaznavanje lahkega vi\u0161avskega je\u010dmena na podlagi izbolj\u0161anega YOLOv5. Plant Methods 21, 42 (2025). https:\/\/doi.org\/10.1186\/s13007-025-01353-0<\/a><\/p>","protected":false},"excerpt":{"rendered":"

\u0160kotski je\u010dmen, odporna \u017eitna kultura, ki jo gojijo v visokogorskih predelih kitajske planote Qinghai-Tibet, igra klju\u010dno vlogo pri lokalni prehranski varnosti in gospodarskem ...<\/p>","protected":false},"author":210249433,"featured_media":11562,"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":"","content-type":"","_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_feature_clip_id":0,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"{title}\n\n{excerpt}\n\n{url}","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2},"_wpas_customize_per_network":false,"jetpack_post_was_ever_published":false},"categories":[1657,1660,1377],"tags":[],"class_list":["post-11559","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-precision-farming","category-agriculture-mapping","category-crop-monitoring"],"acf":[],"yoast_head":"\nBarley Farming Gets a Boost With Lightweight YOLOv5 Detection - GeoPard Agriculture<\/title>\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\/sl\/blog\/gojenje-jecmena-dobi-zagon-z-lahkim-zaznavanjem-yolov5\/\" \/>\n<meta property=\"og:locale\" content=\"sl_SI\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Barley Farming Gets a Boost With Lightweight YOLOv5 Detection - GeoPard Agriculture\" \/>\n<meta property=\"og:description\" content=\"Highland barley, a resilient cereal crop grown in the high-altitude regions of China\u2019s Qinghai-Tibet Plateau, plays a critical role in local food security and economic...\" \/>\n<meta property=\"og:url\" content=\"https:\/\/geopard.tech\/sl\/blog\/gojenje-jecmena-dobi-zagon-z-lahkim-zaznavanjem-yolov5\/\" \/>\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=\"2025-05-04T20:58:53+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2025-05-04T21:00:29+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/geopard.tech\/wp-content\/uploads\/2025\/05\/Barley-Farming-Gets-a-Boost-With-Lightweight-YOLOv5-Detection-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=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Muhammad Farjad\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"8 minut\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\\\/\"},\"author\":{\"name\":\"Muhammad Farjad\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/#\\\/schema\\\/person\\\/123c5562fb47aa8cf3aa81ae91e5e935\"},\"headline\":\"Barley Farming Gets a Boost With Lightweight YOLOv5 Detection\",\"datePublished\":\"2025-05-04T20:58:53+00:00\",\"dateModified\":\"2025-05-04T21:00:29+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\\\/\"},\"wordCount\":1570,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2025\\\/05\\\/Barley-Farming-Gets-a-Boost-With-Lightweight-YOLOv5-Detection.png?fit=3120%2C1755&ssl=1\",\"articleSection\":[\"Precision Farming\",\"Agriculture Mapping\",\"Crop monitoring\"],\"inLanguage\":\"sl-SI\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/geopard.tech\\\/blog\\\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\\\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\\\/\",\"url\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\\\/\",\"name\":\"Barley Farming Gets a Boost With Lightweight YOLOv5 Detection - GeoPard Agriculture\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2025\\\/05\\\/Barley-Farming-Gets-a-Boost-With-Lightweight-YOLOv5-Detection.png?fit=3120%2C1755&ssl=1\",\"datePublished\":\"2025-05-04T20:58:53+00:00\",\"dateModified\":\"2025-05-04T21:00:29+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\\\/#breadcrumb\"},\"inLanguage\":\"sl-SI\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/geopard.tech\\\/blog\\\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"sl-SI\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\\\/#primaryimage\",\"url\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2025\\\/05\\\/Barley-Farming-Gets-a-Boost-With-Lightweight-YOLOv5-Detection.png?fit=3120%2C1755&ssl=1\",\"contentUrl\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2025\\\/05\\\/Barley-Farming-Gets-a-Boost-With-Lightweight-YOLOv5-Detection.png?fit=3120%2C1755&ssl=1\",\"width\":3120,\"height\":1755,\"caption\":\"Barley Farming Gets a Boost With Lightweight YOLOv5 Detection\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/geopard.tech\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Barley Farming Gets a Boost With Lightweight YOLOv5 Detection\"}]},{\"@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\":\"sl-SI\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/#organization\",\"name\":\"GeoPard Agriculture\",\"alternateName\":\"GeoPard\",\"url\":\"https:\\\/\\\/geopard.tech\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"sl-SI\",\"@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\":\"sl-SI\",\"@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 plugin. -->","yoast_head_json":{"title":"Gojenje je\u010dmena dobi spodbudo z lahkim zaznavanjem YOLOv5 - GeoPard Agriculture","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\/sl\/blog\/gojenje-jecmena-dobi-zagon-z-lahkim-zaznavanjem-yolov5\/","og_locale":"sl_SI","og_type":"article","og_title":"Barley Farming Gets a Boost With Lightweight YOLOv5 Detection - GeoPard Agriculture","og_description":"Highland barley, a resilient cereal crop grown in the high-altitude regions of China\u2019s Qinghai-Tibet Plateau, plays a critical role in local food security and economic...","og_url":"https:\/\/geopard.tech\/sl\/blog\/gojenje-jecmena-dobi-zagon-z-lahkim-zaznavanjem-yolov5\/","og_site_name":"GeoPard - Precision agriculture Mapping software","article_publisher":"https:\/\/www.facebook.com\/geopardAgriculture\/","article_published_time":"2025-05-04T20:58:53+00:00","article_modified_time":"2025-05-04T21:00:29+00:00","og_image":[{"width":1024,"height":576,"url":"https:\/\/geopard.tech\/wp-content\/uploads\/2025\/05\/Barley-Farming-Gets-a-Boost-With-Lightweight-YOLOv5-Detection-1024x576.png","type":"image\/png"}],"author":"Muhammad Farjad","twitter_card":"summary_large_image","twitter_creator":"@geopardagri","twitter_site":"@geopardagri","twitter_misc":{"Written by":"Muhammad Farjad","Est. reading time":"8 minut"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/geopard.tech\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/#article","isPartOf":{"@id":"https:\/\/geopard.tech\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/"},"author":{"name":"Muhammad Farjad","@id":"https:\/\/geopard.tech\/#\/schema\/person\/123c5562fb47aa8cf3aa81ae91e5e935"},"headline":"Barley Farming Gets a Boost With Lightweight YOLOv5 Detection","datePublished":"2025-05-04T20:58:53+00:00","dateModified":"2025-05-04T21:00:29+00:00","mainEntityOfPage":{"@id":"https:\/\/geopard.tech\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/"},"wordCount":1570,"commentCount":0,"publisher":{"@id":"https:\/\/geopard.tech\/#organization"},"image":{"@id":"https:\/\/geopard.tech\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/#primaryimage"},"thumbnailUrl":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2025\/05\/Barley-Farming-Gets-a-Boost-With-Lightweight-YOLOv5-Detection.png?fit=3120%2C1755&ssl=1","articleSection":["Precision Farming","Agriculture Mapping","Crop monitoring"],"inLanguage":"sl-SI","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/geopard.tech\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/geopard.tech\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/","url":"https:\/\/geopard.tech\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/","name":"Gojenje je\u010dmena dobi spodbudo z lahkim zaznavanjem YOLOv5 - GeoPard Agriculture","isPartOf":{"@id":"https:\/\/geopard.tech\/#website"},"primaryImageOfPage":{"@id":"https:\/\/geopard.tech\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/#primaryimage"},"image":{"@id":"https:\/\/geopard.tech\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/#primaryimage"},"thumbnailUrl":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2025\/05\/Barley-Farming-Gets-a-Boost-With-Lightweight-YOLOv5-Detection.png?fit=3120%2C1755&ssl=1","datePublished":"2025-05-04T20:58:53+00:00","dateModified":"2025-05-04T21:00:29+00:00","breadcrumb":{"@id":"https:\/\/geopard.tech\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/#breadcrumb"},"inLanguage":"sl-SI","potentialAction":[{"@type":"ReadAction","target":["https:\/\/geopard.tech\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/"]}]},{"@type":"ImageObject","inLanguage":"sl-SI","@id":"https:\/\/geopard.tech\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/#primaryimage","url":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2025\/05\/Barley-Farming-Gets-a-Boost-With-Lightweight-YOLOv5-Detection.png?fit=3120%2C1755&ssl=1","contentUrl":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2025\/05\/Barley-Farming-Gets-a-Boost-With-Lightweight-YOLOv5-Detection.png?fit=3120%2C1755&ssl=1","width":3120,"height":1755,"caption":"Barley Farming Gets a Boost With Lightweight YOLOv5 Detection"},{"@type":"BreadcrumbList","@id":"https:\/\/geopard.tech\/blog\/barley-farming-gets-a-boost-with-lightweight-yolov5-detection\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/geopard.tech\/"},{"@type":"ListItem","position":2,"name":"Barley Farming Gets a Boost With Lightweight YOLOv5 Detection"}]},{"@type":"WebSite","@id":"https:\/\/geopard.tech\/#website","url":"https:\/\/geopard.tech\/","name":"GeoPard - programska oprema za precizno kmetijstvo","description":"Programska oprema za kartiranje natan\u010dnega kmetijstva","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":"sl-SI"},{"@type":"Organization","@id":"https:\/\/geopard.tech\/#organization","name":"GeoPard kmetijstvo","alternateName":"GeoPard","url":"https:\/\/geopard.tech\/","logo":{"@type":"ImageObject","inLanguage":"sl-SI","@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":"Muhamed Farjad","image":{"@type":"ImageObject","inLanguage":"sl-SI","@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\/2025\/05\/Barley-Farming-Gets-a-Boost-With-Lightweight-YOLOv5-Detection.png?fit=3120%2C1755&ssl=1","jetpack_likes_enabled":true,"jetpack_sharing_enabled":true,"jetpack_shortlink":"https:\/\/wp.me\/pdiCPa-30r","_links":{"self":[{"href":"https:\/\/geopard.tech\/sl\/wp-json\/wp\/v2\/posts\/11559","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/geopard.tech\/sl\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/geopard.tech\/sl\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/geopard.tech\/sl\/wp-json\/wp\/v2\/users\/210249433"}],"replies":[{"embeddable":true,"href":"https:\/\/geopard.tech\/sl\/wp-json\/wp\/v2\/comments?post=11559"}],"version-history":[{"count":0,"href":"https:\/\/geopard.tech\/sl\/wp-json\/wp\/v2\/posts\/11559\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/geopard.tech\/sl\/wp-json\/wp\/v2\/media\/11562"}],"wp:attachment":[{"href":"https:\/\/geopard.tech\/sl\/wp-json\/wp\/v2\/media?parent=11559"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/geopard.tech\/sl\/wp-json\/wp\/v2\/categories?post=11559"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/geopard.tech\/sl\/wp-json\/wp\/v2\/tags?post=11559"}],"curies":[{"name":"delovni list","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}