{"id":11618,"date":"2025-05-25T23:15:55","date_gmt":"2025-05-25T21:15:55","guid":{"rendered":"https:\/\/geopard.tech\/?p=11618"},"modified":"2025-05-25T23:15:55","modified_gmt":"2025-05-25T21:15:55","slug":"augstas-precizitates-maksliga-intelekta-modeli-klasifice-topografisko-kartesanu-atrak-neka-tradicionalie-modeli","status":"publish","type":"post","link":"https:\/\/geopard.tech\/lv\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/","title":{"rendered":"Augstas precizit\u0101tes m\u0101ksl\u012bg\u0101 intelekta mode\u013ci klasific\u0113 topogr\u0101fisko kart\u0113\u0161anu \u0101tr\u0101k nek\u0101 tradicion\u0101l\u0101s metodes"},"content":{"rendered":"

Indon\u0113zija, valsts ar vair\u0101k nek\u0101 17\u00a0000 sal\u0101m, kas aiz\u0146em 1,9 miljonus kvadr\u0101tkilometru, saskaras ar b\u016btisku izaicin\u0101jumu detaliz\u0113tu kar\u0161u izveid\u0113, lai atbalst\u012btu savus att\u012bst\u012bbas m\u0113r\u0137us.<\/p>\n

T\u0101 k\u0101 tikai 3% valsts teritorijas ir p\u0101rkl\u0101tas ar liela m\u0113roga topogr\u0101fiskaj\u0101m kart\u0113m (m\u0113rogs 1:5000), tradicion\u0101l\u0101s metodes, piem\u0113ram, manu\u0101la stereogr\u0101fija un lauka apsekojumi, ir p\u0101r\u0101k l\u0113nas, lai apmierin\u0101tu steidzam\u0101s pils\u0113tpl\u0101no\u0161anas, katastrofu p\u0101rvald\u012bbas un vides aizsardz\u012bbas vajadz\u012bbas.<\/p>\n

Novatorisks p\u0113t\u012bjums, kas public\u0113ts\u00a0T\u0101lizp\u0113te<\/em> 2025. gad\u0101 pied\u0101v\u0101 risin\u0101jumu: dzi\u013c\u0101s m\u0101c\u012b\u0161an\u0101s sist\u0113mu, kas automatiz\u0113 zemes seguma klasifik\u0101ciju, izmantojot \u013coti augstas iz\u0161\u0137irtsp\u0113jas satel\u012btatt\u0113lus.<\/p>\n

Indon\u0113zijas kart\u0113\u0161anas izaicin\u0101jums Topogr\u0101fija<\/strong><\/h2>\n

Indon\u0113zijas lielums un sare\u017e\u0123\u012bt\u012bba padara kart\u0113\u0161anu par milz\u012bgu uzdevumu. \u0122eotelpisk\u0101s inform\u0101cijas a\u0123ent\u016bra (BIG), kas ir atbild\u012bga par valsts kart\u0113\u0161anu, pa\u0161laik katru gadu sara\u017eo 13\u00a0000 kvadr\u0101tkilometru topogr\u0101fisko kar\u0161u.<\/p>\n

\u0160\u0101d\u0101 temp\u0101 visas valsts kart\u0113\u0161ana pras\u012btu vair\u0101k nek\u0101 gadsimtu. Pat ja neskaita me\u017eu teritorijas, kas kl\u0101j gandr\u012bz pusi Indon\u0113zijas, atliku\u0161\u0101s teritorijas pabeig\u0161ana joproj\u0101m pras\u012btu 60 gadus.<\/p>\n

\u0160is l\u0113nais progress ir pretrun\u0101 ar nacion\u0101laj\u0101m priorit\u0101t\u0113m, piem\u0113ram,\u00a0Vienas kartes politika<\/em>, kas tika ieviesta 2016. gad\u0101, lai standartiz\u0113tu kartes da\u017e\u0101d\u0101s nozar\u0113s un izvair\u012btos no konfliktiem zemes izmanto\u0161an\u0101. \u0160\u012bs politikas piel\u0101go\u0161ana 1:5000 kart\u0113m ir b\u016btiska, ta\u010du t\u0101 krietni atpaliek no grafika.<\/p>\n

Topogr\u0101fisk\u0101s kartes<\/strong>\u00a0ir detaliz\u0113ti Zemes virsmas dabisko un cilv\u0113ka rad\u012bto veidojumu att\u0113lojumi, tostarp augstumi (pakalni, ielejas), \u016bdenstilpes, ce\u013ci, \u0113kas un ve\u0123et\u0101cija.<\/p>\n

Tie kalpo k\u0101 pamatinstrumenti infrastrukt\u016bras pl\u0101no\u0161anai, katastrofu seku likvid\u0113\u0161anai un vides monitoringam. Indon\u0113zij\u0101 \u0161o kar\u0161u izveide m\u0113rog\u0101 1:5000 (kur 1 cm kart\u0113 atbilst 50 metriem uz zemes) ir kritiski svar\u012bga precizit\u0101tei t\u0101dos projektos k\u0101 ce\u013cu b\u016bvniec\u012bba vai pl\u016bdu model\u0113\u0161ana.<\/p>\n

\"Indon\u0113zijas<\/p>\n

Zemes seguma dati<\/strong>, topogr\u0101fisko kar\u0161u apak\u0161kopa, attiecas uz fizisko materi\u0101lu uz Zemes virsmas, piem\u0113ram, me\u017eiem, pils\u0113tu teritorij\u0101m vai \u016bdeni. At\u0161\u0137ir\u012bb\u0101 no\u00a0zemes izmanto\u0161ana<\/em>\u00a0(kas apraksta, k\u0101 cilv\u0113ki izmanto zemi, piem\u0113ram, dz\u012bvojam\u0101s vai r\u016bpniecisk\u0101s zonas), zeme koncentr\u0113jas uz nov\u0113rojam\u0101m iez\u012bm\u0113m.<\/p>\n

Prec\u012bzas zemes seguma kartes pal\u012bdz vald\u012bb\u0101m izsekot me\u017eu izcir\u0161anai, uzraudz\u012bt pils\u0113tu izple\u0161anos vai nov\u0113rt\u0113t lauksaimniec\u012bbas produktivit\u0101ti. Tradicion\u0101li anal\u012bti\u0137i \u0161os objektus manu\u0101li mar\u0137\u0113 pikseli pa pikselim, izmantojot aerofotogr\u0101fijas vai satel\u012btatt\u0113lus, un \u0161is process ir gan laikietilp\u012bgs, gan pak\u013cauts cilv\u0113cisk\u0101m k\u013c\u016bd\u0101m.<\/p>\n

Piem\u0113ram, ce\u013cu vai mazu \u0113ku identific\u0113\u0161ana bl\u012bvi apdz\u012bvot\u0101s pils\u0113tu teritorij\u0101s var pras\u012bt vair\u0101kas dienas r\u016bp\u012bga darba. 2025.\u00a0gada p\u0113t\u012bjum\u0101 \u0161\u012b probl\u0113ma tiek risin\u0101ta, aizst\u0101jot manu\u0101los centienus ar m\u0101ksl\u012bgo intelektu, \u012bpa\u0161i dzi\u013co m\u0101c\u012b\u0161anos, lai automatiz\u0113tu zemes seguma klasifik\u0101ciju.<\/p>\n

M\u0101ksl\u012bg\u0101 intelekta vad\u012bta satel\u012btatt\u0113lu anal\u012bze\u00a0<\/strong><\/h2>\n

P\u0113t\u012bjums k\u0101 testa gad\u012bjums bija v\u0113rsts uz Mataramas pils\u0113tu, nelielu, bet daudzveid\u012bgu pils\u0113tas teritoriju Lombokas sal\u0101. Komanda izmantoja\u00a0Plej\u0101du satel\u012btatt\u0113li<\/strong>\u00a0no 2015.\u00a0gada, kur\u0101 bija iek\u013cauti augstas iz\u0161\u0137irtsp\u0113jas panhromatiski (0,5\u00a0metri) un multispektr\u0101li (2\u00a0metri) dati.<\/p>\n

Panhromatiskie att\u0113li uztver smalkas telpiskas deta\u013cas pel\u0113kto\u0146os, savuk\u0101rt multispektr\u0101lie att\u0113li sniedz kr\u0101su un infrasarkano staru inform\u0101ciju noteiktos vi\u013c\u0146u garuma diapazonos (piem\u0113ram, sarkan\u0101, za\u013c\u0101, zil\u0101, tuv\u0101 infrasarkan\u0101).<\/p>\n

Lai apvienotu \u0161\u012bs stipr\u0101s puses, p\u0113tnieki izmantoja pan-sharpening metodi, kas apvieno augstas iz\u0161\u0137irtsp\u0113jas pel\u0113kto\u0146u datus ar zem\u0101kas iz\u0161\u0137irtsp\u0113jas kr\u0101su att\u0113liem. \u0160is process rad\u012bja asus, detaliz\u0113tus att\u0113lus ar 0,5 metru iz\u0161\u0137irtsp\u0113ju, kas ir ide\u0101li piem\u0113roti mazu objektu, piem\u0113ram, ce\u013cu vai atsevi\u0161\u0137u \u0113ku, noteik\u0161anai.<\/p>\n

Pan-asin\u0101\u0161ana ir \u013coti svar\u012bga, jo t\u0101 saglab\u0101 bag\u0101t\u012bgo multispektr\u0101lo datu spektr\u0101lo inform\u0101ciju, vienlaikus uzlabojot telpisko skaidr\u012bbu, nodro\u0161inot, ka kr\u0101sas prec\u012bzi atbilst fiziskaj\u0101m iez\u012bm\u0113m.<\/p>\n

P\u0113c tam komanda no att\u0113liem ieguva papildu inform\u0101ciju, lai uzlabotu klasifik\u0101cijas precizit\u0101ti. Vi\u0146i apr\u0113\u0137in\u0101ja normaliz\u0113to ve\u0123et\u0101cijas diferenci\u0101lo indeksu (NDVI), kas ir augu vesel\u012bbas r\u0101d\u012bt\u0101js, kas ieg\u016bts no tuv\u0101 infrasarkan\u0101 starojuma (NIR) un sarkan\u0101s gaismas atstaro\u0161anas.<\/p>\n

Vesel\u012bga ve\u0123et\u0101cija atstaro vair\u0101k tuv\u0101 infrasarkan\u0101 starojuma un absorb\u0113 vair\u0101k sarkan\u0101s gaismas hlorofila aktivit\u0101tes d\u0113\u013c. Formula\u00a0NDVI = (NIR \u2212 Sarkans) \/ (NIR + Sarkans)<\/span>NDVI<\/span><\/span>=<\/span><\/span>(<\/span>tuv\u0101 infrasarkan\u0101 starojuma (NIR)<\/span><\/span>\u2212<\/span><\/span>Sarkans<\/span><\/span>)<\/span>\/<\/span>(<\/span>tuv\u0101 infrasarkan\u0101 starojuma (NIR)<\/span><\/span>+<\/span><\/span>Sarkans<\/span><\/span>)<\/span><\/span><\/span><\/span>\u00a0ieg\u016bst v\u0113rt\u012bbas no -1 l\u012bdz 1, kur augst\u0101kas v\u0113rt\u012bbas nor\u0101da uz bl\u012bv\u0101ku, vesel\u012bg\u0101ku ve\u0123et\u0101ciju.<\/p>\n

NDVI ir nenov\u0113rt\u0113jams, lai at\u0161\u0137irtu me\u017eus, lauksaimniec\u012bbas zemi un pils\u0113tu za\u013c\u0101s zonas. Piem\u0113ram, \u0161aj\u0101 p\u0113t\u012bjum\u0101 NDVI pal\u012bdz\u0113ja at\u0161\u0137irt sul\u012bgas plant\u0101cijas no kailas augsnes.<\/p>\n

Tekst\u016bras anal\u012bze<\/strong>\u00a0bija v\u0113l viens svar\u012bgs solis. Izmantojot statistisko metodi, ko sauc par pel\u0113k\u0101 l\u012bme\u0146a l\u012bdz\u0101spar\u0101d\u012b\u0161an\u0101s matricu (GLCM), p\u0113tnieki kvantific\u0113ja att\u0113los redzamos mode\u013cus, piem\u0113ram, lauksaimniec\u012bbas lauku nel\u012bdzenumu sal\u012bdzin\u0101jum\u0101 ar bru\u0123\u0113to ce\u013cu gludumu.<\/p>\n

GLCM darbojas, analiz\u0113jot, cik bie\u017ei att\u0113l\u0101 par\u0101d\u0101s pikse\u013cu p\u0101ri ar noteikt\u0101m v\u0113rt\u012bb\u0101m un telpisk\u0101m attiec\u012bb\u0101m (piem\u0113ram, horizont\u0101li blakus). No \u0161\u012bs matricas var ieg\u016bt t\u0101dus r\u0101d\u012bt\u0101jus k\u0101\u00a0viendab\u012bgums<\/em>\u00a0(pikse\u013cu v\u0113rt\u012bbu vienveid\u012bba),\u00a0kontrasts<\/em>\u00a0(lok\u0101l\u0101s intensit\u0101tes sv\u0101rst\u012bbas) un\u00a0entropija<\/em>\u00a0(pikse\u013cu sadal\u012bjuma nejau\u0161\u012bba) tiek apr\u0113\u0137in\u0101ti.<\/p>\n

\u0160ie tekst\u016bras r\u0101d\u012bt\u0101ji pal\u012bdz\u0113ja m\u0101ksl\u012bg\u0101 intelekta modelim at\u0161\u0137irt l\u012bdz\u012bga izskata zemes seguma veidus, piem\u0113ram, at\u0161\u0137irot asfalta ce\u013cus no tum\u0161iem augsnes ple\u0137iem.<\/p>\n

Lai vienk\u0101r\u0161otu datus, komanda pielietoja\u00a0Galveno komponentu anal\u012bze (PCA)<\/strong>, metode, kas identific\u0113 noz\u012bm\u012bg\u0101kos mode\u013cus datu kop\u0101. PCA samazina redundanci, p\u0101rveidojot korel\u0113tos main\u012bgos (piem\u0113ram, vair\u0101kas tekst\u016bras joslas) maz\u0101k\u0101 nekorel\u0113tu komponentu kop\u0101.<\/p>\n

\u0160aj\u0101 p\u0113t\u012bjum\u0101 PCA sa\u012bsin\u0101ja piecas tekst\u016bras joslas div\u0101s galvenaj\u0101s komponent\u0113s, vienlaikus saglab\u0101jot s\u0101kotn\u0113j\u0101s inform\u0101cijas 95%. Tas racionaliz\u0113ja dzi\u013c\u0101s m\u0101c\u012b\u0161an\u0101s mode\u013ca ievadi, uzlabojot gan precizit\u0101ti, gan skait\u013co\u0161anas efektivit\u0101ti.<\/p>\n

U-Net dzi\u013c\u0101 m\u0101c\u012b\u0161an\u0101s zemes seguma izp\u0113tei <\/strong>Topogr\u0101fija<\/strong><\/h2>\n

P\u0113t\u012bjuma pamat\u0101 bija dzi\u013c\u0101s m\u0101c\u012b\u0161an\u0101s modelis, kas balst\u012bts uz U-Net arhitekt\u016bru \u2014 konvolucion\u0101l\u0101 neironu t\u012bkla (CNN) veidu, ko pla\u0161i izmanto att\u0113lu segment\u0101cijas uzdevumos.<\/p>\n

U-Net, kas nosaukts U-veida dizaina d\u0113\u013c, sast\u0101v no div\u0101m galvenaj\u0101m da\u013c\u0101m: kod\u0113t\u0101ja, kas analiz\u0113 att\u0113lu, lai ieg\u016btu hierarhiskas paz\u012bmes (piem\u0113ram, malas, tekst\u016bras), un dekod\u0113t\u0101ja, kas rekonstru\u0113 att\u0113lu ar pikse\u013ciem atbilsto\u0161\u0101m eti\u0137et\u0113m.<\/p>\n

Kod\u0113t\u0101js izmanto konvolucion\u0101los sl\u0101\u0146us un apvieno\u0161anu, lai samazin\u0101tu att\u0113la iz\u0161\u0137irtsp\u0113ju, uztverot pla\u0161us mode\u013cus, savuk\u0101rt dekod\u0113t\u0101js palielina datu iz\u0161\u0137irtsp\u0113ju, lai atjaunotu telpisko iz\u0161\u0137irtsp\u0113ju. Savienojumu izlai\u0161ana starp kod\u0113t\u0101ja un dekod\u0113t\u0101ja sl\u0101\u0146iem saglab\u0101 smalkas deta\u013cas, nodro\u0161inot prec\u012bzu robe\u017eu noteik\u0161anu, kas ir kritiski svar\u012bga funkcija \u0161auru ce\u013cu vai neregul\u0101ras formas \u0113ku kart\u0113\u0161anai.<\/p>\n

\"Zemes<\/p>\n

Model\u012b tika izmantots ResNet34 mugurkauls \u2014 iepriek\u0161 apm\u0101c\u012bts t\u012bkls, kas paz\u012bstams ar savu dzi\u013cumu un efektivit\u0101ti. ResNet34 pieder pie atliku\u0161o t\u012bklu saimes, kas ievie\u0161 "\u012bsce\u013cu savienojumus", lai apietu sl\u0101\u0146us, mazinot izz\u016bdo\u0161\u0101 gradienta probl\u0113mu (kur dzi\u013cajiem t\u012bkliem ir gr\u016bt\u012bbas m\u0101c\u012bties samazin\u0101to atjaunin\u0101jumu d\u0113\u013c apm\u0101c\u012bbas laik\u0101).<\/p>\n

Izmantojot ResNet34 sp\u0113ju atpaz\u012bt sare\u017e\u0123\u012btus mode\u013cus no ImageNet (masveida att\u0113lu datub\u0101zes), modelim bija nepiecie\u0161ams maz\u0101k apm\u0101c\u012bbas datu un laika, lai piel\u0101gotos satel\u012btatt\u0113liem.<\/p>\n

Mode\u013ca apm\u0101c\u012bbai bija nepiecie\u0161amas 1440 att\u0113lu fl\u012bzes, katra 512 \u00d7 512 pikse\u013ci, aptverot se\u0161as zemes seguma klases: \u0113kas, ce\u013cus, lauksaimniec\u012bbas zemi, kailo zemi, plant\u0101cijas un \u016bdenstilpes.<\/p>\n

Datu kopai bija rakstur\u012bga nel\u012bdzsvarot\u012bba; ce\u013ci un \u016bdenstilpes veidoja attiec\u012bgi tikai 3,7% un 4,2% no izlas\u0113m, savuk\u0101rt \u0113kas un lauksaimniec\u012bbas zeme veidoja vair\u0101k nek\u0101 25% katra. Neskatoties uz \u0161o izaicin\u0101jumu, modelis tika apm\u0101c\u012bts 200 epohos \u2014 l\u012bdzsvars starp precizit\u0101ti un skait\u013co\u0161anas izmaks\u0101m \u2014 ar partijas lielumu 2 atmi\u0146as ierobe\u017eojumu d\u0113\u013c.<\/p>\n

Un\u00a0laikmets<\/strong>\u00a0attiecas uz vienu piln\u012bgu apm\u0101c\u012bbas datu nodo\u0161anu caur modeli, savuk\u0101rt\u00a0partijas lielums<\/strong>\u00a0nosaka, cik paraugu tiek apstr\u0101d\u0101ti pirms mode\u013ca parametru atjaunin\u0101\u0161anas. Maz\u0101ki partijas lielumi samazina atmi\u0146as izmanto\u0161anu, bet var pal\u0113nin\u0101t apm\u0101c\u012bbu.<\/p>\n

Kar\u0161u uzlabo\u0161ana ar morfolo\u0123isko apstr\u0101di<\/strong><\/h2>\n

Pat lab\u0101kie m\u0101ksl\u012bg\u0101 intelekta mode\u013ci rada k\u013c\u016bdas, piem\u0113ram, atsevi\u0161\u0137u pikse\u013cu nepareizu klasific\u0113\u0161anu vai robainu malu veido\u0161anos ap elementiem. Lai to nov\u0113rstu, p\u0113tnieki izmantoja morfolo\u0123isko apstr\u0101di \u2014 metodi, kas izl\u012bdzina nepiln\u012bbas, izmantojot t\u0101das darb\u012bbas k\u0101 erozija un dilat\u0101cija.<\/p>\n

Erozija no\u0146em pl\u0101nus pikse\u013cu sl\u0101\u0146us no objektu robe\u017e\u0101m, likvid\u0113jot s\u012bkus, nepareizi klasific\u0113tus laukumus, savuk\u0101rt dilat\u0101cija pievieno pikse\u013cus, lai papla\u0161in\u0101tu objektu robe\u017eas, aizpildot tuk\u0161umus line\u0101ros objektos, piem\u0113ram, ce\u013cos.<\/p>\n

\u0160\u012bs darb\u012bbas balst\u0101s uz struktur\u0113jo\u0161u elementu (nelielu matricu), kas sl\u012bd p\u0101ri att\u0113lam, lai modific\u0113tu pikse\u013cu v\u0113rt\u012bbas. Optim\u0101lais kodola izm\u0113rs \u0161\u012bm darb\u012bb\u0101m (5 \u00d7 5 pikse\u013ci) tika noteikts, izmantojot pusvarian\u010du anal\u012bzi \u2014 \u0123eostatistisko metodi, kas kvantific\u0113ja telpiskos mode\u013cus att\u0113los.<\/p>\n

Pusvari\u0101cija m\u0113ra, cik daudz pikse\u013cu v\u0113rt\u012bbas at\u0161\u0137iras da\u017e\u0101dos att\u0101lumos, pal\u012bdzot noteikt m\u0113rogu, kur\u0101 tekst\u016bras elementi (piem\u0113ram, \u0113ku kopas) ir visizteikt\u0101kie.<\/p>\n

M\u0101ksl\u012bgais intelekts palielina kart\u0113\u0161anas \u0101trumu un precizit\u0101ti<\/strong><\/h2>\n

Modelis sasniedza s\u0101kotn\u0113jo precizit\u0101ti 84% (kappa r\u0101d\u012bt\u0101js<\/strong>\u00a0= 0,79), kas p\u0113c p\u0113capstr\u0101des pieauga l\u012bdz 86% (kappa = 0,81).\u00a0kappa r\u0101d\u012bt\u0101js<\/strong>\u00a0(Koena kappa) m\u0113ra atbilst\u012bbu starp prognoz\u0113to un faktisko klasifik\u0101ciju, kori\u0123\u0113jot nejau\u0161\u012bbu.<\/p>\n

Rezult\u0101ts 0,81 nor\u0101da uz \u201cgandr\u012bz piln\u012bgu\u201d atbilst\u012bbu, p\u0101rsniedzot 0,61\u20130,80 diapazonu, kas tiek uzskat\u012bts par \u201cb\u016btisku\u201d. \u016adenstilpes un st\u0101d\u012bjumi tika klasific\u0113ti ar gandr\u012bz piln\u012bgu precizit\u0101ti (attiec\u012bgi 97% un 96%), savuk\u0101rt ce\u013ci, kuriem bija rakstur\u012bga pl\u0101na, line\u0101ra forma un \u0113nas, sasniedza 85%.<\/p>\n

\"M\u0101ksl\u012bgais<\/p>\n

Ar\u012b \u0113kas un lauksaimniec\u012bbas zeme uzr\u0101d\u012bja labus rezult\u0101tus, ieg\u016bstot F1 r\u0101d\u012bt\u0101jus 88% un 83%. F1 r\u0101d\u012bt\u0101js, kas ir precizit\u0101tes un atcer\u0113\u0161an\u0101s sp\u0113jas harmoniskais vid\u0113jais r\u0101d\u012bt\u0101js, l\u012bdzsvaro k\u013c\u016bdaini pozit\u012bvus un k\u013c\u016bdaini negat\u012bvus rezult\u0101tus, padarot to ide\u0101li piem\u0113rotu nel\u012bdzsvarotu datu kopu nov\u0113rt\u0113\u0161anai.<\/p>\n

Efektivit\u0101tes pieaugums bija v\u0113l p\u0101rsteidzo\u0161\u0101ks. Tradicion\u0101l\u0101 stereogr\u0101f\u0113\u0161ana, kas ietver elementu manu\u0101lu mar\u0137\u0113\u0161anu 3D aeroatt\u0113los, \u0113ku un ve\u0123et\u0101cijas att\u0113lo\u0161anai aiz\u0146em devi\u0146as dienas uz vienu kartes lapu (5,29 km\u00b2).<\/p>\n

M\u0101ksl\u012bg\u0101 intelekta vad\u012bt\u0101 pieeja samazin\u0101ja \u0161o laiku l\u012bdz 43 min\u016bt\u0113m katrai lapai, kas ir 250 rei\u017eu uzlabojums. Mode\u013ca apm\u0101c\u012bbai s\u0101kotn\u0113ji bija nepiecie\u0161amas 17 stundas, bet p\u0113c apm\u0101c\u012bbas tas var\u0113tu klasific\u0113t pla\u0161as teritorijas ar minim\u0101lu cilv\u0113ka iejauk\u0161anos. \u0160\u012bs sist\u0113mas m\u0113rogo\u0161ana \u013cautu Indon\u0113zijai kart\u0113t 9000 km\u00b2 gad\u0101, samazinot prognoz\u0113to pabeig\u0161anas laiku no vair\u0101k nek\u0101 gadsimta l\u012bdz tikai 15 gadiem.<\/p>\n

M\u0101ksl\u012bg\u0101 intelekta kart\u0113\u0161ana veicina glob\u0101lo ilgtsp\u0113j\u012bbu<\/strong><\/h2>\n

Sekas sniedzas t\u0101lu aiz Indon\u0113zijas robe\u017e\u0101m. Automatiz\u0113ta zemes seguma klasifik\u0101cija atbalsta glob\u0101lus centienus, piem\u0113ram, ANO ilgtsp\u0113j\u012bgas att\u012bst\u012bbas m\u0113r\u0137us (IAM). Piem\u0113ram, me\u017eu izcir\u0161anas (IAM 15) vai pils\u0113tu izple\u0161an\u0101s (IAM 11) izseko\u0161ana k\u013c\u016bst \u0101tr\u0101ka un prec\u012bz\u0101ka.<\/p>\n

Katastrofu apdraud\u0113tos re\u0123ionos, piem\u0113ram, pl\u016bdu apdraud\u0113t\u0101s teritorij\u0101s, atjaunin\u0101tas kartes var identific\u0113t neaizsarg\u0101tas kopienas un pl\u0101not evaku\u0101cijas mar\u0161rutus.<\/p>\n

Ar\u012b lauksaimnieki g\u016bst labumu; prec\u012bzi zemes seguma dati \u013cauj veikt prec\u012bzu lauksaimniec\u012bbu, optimiz\u0113jot \u016bdens izmanto\u0161anu un kult\u016braugu ra\u017eu, uzraugot augsnes vesel\u012bbu un ve\u0123et\u0101cijas stresu, izmantojot NDVI.<\/p>\n

Tom\u0113r joproj\u0101m past\u0101v probl\u0113mas. Mode\u013ca veiktsp\u0113ja nepietiekami p\u0101rst\u0101v\u0113t\u0101s klas\u0113s, piem\u0113ram, ce\u013cos, uzsver nepiecie\u0161am\u012bbu p\u0113c l\u012bdzsvarotiem apm\u0101c\u012bbas datiem. Turpm\u0101kaj\u0101 darb\u0101 var\u0113tu iek\u013caut p\u0101rneses m\u0101c\u012b\u0161anos \u2014 metodi, kur\u0101 modelis, kas iepriek\u0161 apm\u0101c\u012bts vienam uzdevumam (piem\u0113ram, visp\u0101r\u0113jai att\u0113lu atpaz\u012b\u0161anai), tiek prec\u012bzi piel\u0101gots konkr\u0113tam pielietojumam (piem\u0113ram, ce\u013cu noteik\u0161anai satel\u012btatt\u0113los).<\/p>\n

Tas samazina nepiecie\u0161am\u012bbu p\u0113c milz\u012bgiem mar\u0137\u0113tiem datu kopumiem, kuru izveide ir d\u0101rga. Precizit\u0101ti var\u0113tu v\u0113l vair\u0101k uzlabot, test\u0113jot t\u0101das progres\u012bvas arhitekt\u016bras k\u0101 U-Net3+, kas uzlabo iez\u012bmju apkopo\u0161anu da\u017e\u0101dos m\u0113rogos, vai uz transformatoriem balst\u012btus mode\u013cus (kas izce\u013cas ar liela att\u0101luma atkar\u012bbu tver\u0161anu att\u0113los).<\/p>\n

Tom\u0113r Lidar (gaismas noteik\u0161anas un diapazona noteik\u0161anas) vai radara datu integr\u0113\u0161ana var\u0113tu ar\u012b uzlabot rezult\u0101tus, \u012bpa\u0161i m\u0101ko\u0146ainos re\u0123ionos, kur optiskajiem satel\u012btiem ir gr\u016bt\u012bbas.<\/p>\n

Secin\u0101jums: Jauna \u0113ra \u0123eotelpiskaj\u0101 zin\u0101tn\u0113<\/h2>\n

\u0160is p\u0113t\u012bjums iez\u012bm\u0113 pagrieziena punktu topogr\u0101fiskaj\u0101 kart\u0113\u0161an\u0101. Automatiz\u0113jot zemes seguma klasifik\u0101ciju, valstis var izveidot prec\u012bzas kartes \u0101tr\u0101k un l\u0113t\u0101k nek\u0101 jebkad agr\u0101k. Indon\u0113zijai \u0161\u012b tehnolo\u0123ija nav tikai \u0113rt\u012bbas \u2014 t\u0101 ir nepiecie\u0161am\u012bba, lai p\u0101rvald\u012btu straujo urbaniz\u0101ciju, aizsarg\u0101tu me\u017eus un sagatavotos ar klimatu saist\u012bt\u0101m katastrof\u0101m.<\/p>\n

Att\u012bstoties m\u0101ksl\u012bgajam intelektam un satel\u012bttehnolo\u0123ij\u0101m, re\u0101llaika, augstas iz\u0161\u0137irtsp\u0113jas kart\u0113\u0161anas v\u012bzija ir sasniedzama, dodot vald\u012bb\u0101m un kopien\u0101m iesp\u0113ju veidot ilgtsp\u0113j\u012bg\u0101ku n\u0101kotni.<\/p>\n

Atsauce<\/strong>Hakim, YF; Tsai, F. Uz dzi\u013c\u0101s m\u0101c\u012b\u0161an\u0101s balst\u012bta zemes seguma ieg\u016b\u0161ana no \u013coti augstas iz\u0161\u0137irtsp\u0113jas satel\u012btatt\u0113liem, lai pal\u012bdz\u0113tu veidot liela m\u0113roga topogr\u0101fisk\u0101s kartes. Remote Sens. 2025, 17, 473. https:\/\/doi.org\/10.3390\/rs17030473<\/a><\/p>","protected":false},"excerpt":{"rendered":"

Indon\u0113zija, valsts ar vair\u0101k nek\u0101 17\u00a0000 sal\u0101m, kas aiz\u0146em 1,9 miljonus kvadr\u0101tkilometru, saskaras ar b\u016btisku izaicin\u0101jumu detaliz\u0113tu kar\u0161u izveid\u0113, lai atbalst\u012btu savus att\u012bst\u012bbas m\u0113r\u0137us\u2026.<\/p>","protected":false},"author":210157960,"featured_media":11626,"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,"_wpcom_ai_launchpad_first_post":false,"_jetpack_hide_featured_image":false,"_jetpack_featured_image_duplicate_dismissed":0,"_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":[1661,1366],"tags":[],"class_list":["post-11618","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-satellite-imagery","category-topography"],"acf":[],"yoast_head":"\nHigh-Accuracy AI Models Classifies Topographic Mapping Faster Than Traditional - GeoPard<\/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\/lv\/emuars\/augstas-precizitates-maksliga-intelekta-modeli-klasifice-topografisko-kartesanu-atrak-neka-tradicionalie-modeli\/\" \/>\n<meta property=\"og:locale\" content=\"lv_LV\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"High-Accuracy AI Models Classifies Topographic Mapping Faster Than Traditional - GeoPard\" \/>\n<meta property=\"og:description\" content=\"Indonesia, a nation of over 17,000 islands spanning 1.9 million square kilometers, faces a critical challenge in creating detailed maps to support its development goals....\" \/>\n<meta property=\"og:url\" content=\"https:\/\/geopard.tech\/lv\/emuars\/augstas-precizitates-maksliga-intelekta-modeli-klasifice-topografisko-kartesanu-atrak-neka-tradicionalie-modeli\/\" \/>\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=\"2025-05-25T21:15:55+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/geopard.tech\/wp-content\/uploads\/2025\/05\/High-Accuracy-AI-Models-Classifies-Topographic-Mapping-Faster-Than-Traditional-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=\"dementievgeopard\" \/>\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=\"dementievgeopard\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"9 min\u016btes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\\\/\"},\"author\":{\"name\":\"dementievgeopard\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/#\\\/schema\\\/person\\\/dd217733c742620adc57befbbcd84a8a\"},\"headline\":\"High-Accuracy AI Models Classifies Topographic Mapping Faster Than Traditional\",\"datePublished\":\"2025-05-25T21:15:55+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\\\/\"},\"wordCount\":1728,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2025\\\/05\\\/High-Accuracy-AI-Models-Classifies-Topographic-Mapping-Faster-Than-Traditional.png?fit=1920%2C1080&ssl=1\",\"articleSection\":[\"Satellite Imagery\",\"Topography\"],\"inLanguage\":\"lv-LV\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/geopard.tech\\\/blog\\\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\\\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\\\/\",\"url\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\\\/\",\"name\":\"High-Accuracy AI Models Classifies Topographic Mapping Faster Than Traditional - GeoPard\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2025\\\/05\\\/High-Accuracy-AI-Models-Classifies-Topographic-Mapping-Faster-Than-Traditional.png?fit=1920%2C1080&ssl=1\",\"datePublished\":\"2025-05-25T21:15:55+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\\\/#breadcrumb\"},\"inLanguage\":\"lv-LV\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/geopard.tech\\\/blog\\\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"lv-LV\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\\\/#primaryimage\",\"url\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2025\\\/05\\\/High-Accuracy-AI-Models-Classifies-Topographic-Mapping-Faster-Than-Traditional.png?fit=1920%2C1080&ssl=1\",\"contentUrl\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2025\\\/05\\\/High-Accuracy-AI-Models-Classifies-Topographic-Mapping-Faster-Than-Traditional.png?fit=1920%2C1080&ssl=1\",\"width\":1920,\"height\":1080,\"caption\":\"High-Accuracy AI Models Classifies Topographic Mapping Faster Than Traditional\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/blog\\\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/geopard.tech\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"High-Accuracy AI Models Classifies Topographic Mapping Faster Than Traditional\"}]},{\"@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\":\"lv-LV\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/#organization\",\"name\":\"GeoPard Agriculture\",\"alternateName\":\"GeoPard\",\"url\":\"https:\\\/\\\/geopard.tech\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"lv-LV\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/geopard_logo_transparentbackground.png?fit=512%2C68&ssl=1\",\"contentUrl\":\"https:\\\/\\\/i0.wp.com\\\/geopard.tech\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/geopard_logo_transparentbackground.png?fit=512%2C68&ssl=1\",\"width\":512,\"height\":68,\"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\\\/\",\"https:\\\/\\\/www.youtube.com\\\/channel\\\/UCiaPGLAhRPNh-s85dXdC-Sw\",\"https:\\\/\\\/www.g2.com\\\/products\\\/geopard-agriculture-precision-farming-software\\\/reviews\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/geopard.tech\\\/#\\\/schema\\\/person\\\/dd217733c742620adc57befbbcd84a8a\",\"name\":\"dementievgeopard\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"lv-LV\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/7c09adac9c20b00454199df5a41ea67c812a6c97e948bf109899836ddde31354?s=96&d=identicon&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/7c09adac9c20b00454199df5a41ea67c812a6c97e948bf109899836ddde31354?s=96&d=identicon&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/7c09adac9c20b00454199df5a41ea67c812a6c97e948bf109899836ddde31354?s=96&d=identicon&r=g\",\"caption\":\"dementievgeopard\"}}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Augstas precizit\u0101tes m\u0101ksl\u012bg\u0101 intelekta mode\u013ci klasific\u0113 topogr\u0101fisko kart\u0113\u0161anu \u0101tr\u0101k nek\u0101 tradicion\u0101l\u0101s \u2014 GeoPard","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\/lv\/emuars\/augstas-precizitates-maksliga-intelekta-modeli-klasifice-topografisko-kartesanu-atrak-neka-tradicionalie-modeli\/","og_locale":"lv_LV","og_type":"article","og_title":"High-Accuracy AI Models Classifies Topographic Mapping Faster Than Traditional - GeoPard","og_description":"Indonesia, a nation of over 17,000 islands spanning 1.9 million square kilometers, faces a critical challenge in creating detailed maps to support its development goals....","og_url":"https:\/\/geopard.tech\/lv\/emuars\/augstas-precizitates-maksliga-intelekta-modeli-klasifice-topografisko-kartesanu-atrak-neka-tradicionalie-modeli\/","og_site_name":"GeoPard - Precision agriculture Mapping software","article_publisher":"https:\/\/www.facebook.com\/geopardAgriculture\/","article_published_time":"2025-05-25T21:15:55+00:00","og_image":[{"width":1024,"height":576,"url":"https:\/\/geopard.tech\/wp-content\/uploads\/2025\/05\/High-Accuracy-AI-Models-Classifies-Topographic-Mapping-Faster-Than-Traditional-1024x576.png","type":"image\/png"}],"author":"dementievgeopard","twitter_card":"summary_large_image","twitter_creator":"@geopardagri","twitter_site":"@geopardagri","twitter_misc":{"Written by":"dementievgeopard","Est. reading time":"9 min\u016btes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/geopard.tech\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/#article","isPartOf":{"@id":"https:\/\/geopard.tech\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/"},"author":{"name":"dementievgeopard","@id":"https:\/\/geopard.tech\/#\/schema\/person\/dd217733c742620adc57befbbcd84a8a"},"headline":"High-Accuracy AI Models Classifies Topographic Mapping Faster Than Traditional","datePublished":"2025-05-25T21:15:55+00:00","mainEntityOfPage":{"@id":"https:\/\/geopard.tech\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/"},"wordCount":1728,"commentCount":0,"publisher":{"@id":"https:\/\/geopard.tech\/#organization"},"image":{"@id":"https:\/\/geopard.tech\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/#primaryimage"},"thumbnailUrl":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2025\/05\/High-Accuracy-AI-Models-Classifies-Topographic-Mapping-Faster-Than-Traditional.png?fit=1920%2C1080&ssl=1","articleSection":["Satellite Imagery","Topography"],"inLanguage":"lv-LV","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/geopard.tech\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/geopard.tech\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/","url":"https:\/\/geopard.tech\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/","name":"Augstas precizit\u0101tes m\u0101ksl\u012bg\u0101 intelekta mode\u013ci klasific\u0113 topogr\u0101fisko kart\u0113\u0161anu \u0101tr\u0101k nek\u0101 tradicion\u0101l\u0101s \u2014 GeoPard","isPartOf":{"@id":"https:\/\/geopard.tech\/#website"},"primaryImageOfPage":{"@id":"https:\/\/geopard.tech\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/#primaryimage"},"image":{"@id":"https:\/\/geopard.tech\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/#primaryimage"},"thumbnailUrl":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2025\/05\/High-Accuracy-AI-Models-Classifies-Topographic-Mapping-Faster-Than-Traditional.png?fit=1920%2C1080&ssl=1","datePublished":"2025-05-25T21:15:55+00:00","breadcrumb":{"@id":"https:\/\/geopard.tech\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/#breadcrumb"},"inLanguage":"lv-LV","potentialAction":[{"@type":"ReadAction","target":["https:\/\/geopard.tech\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/"]}]},{"@type":"ImageObject","inLanguage":"lv-LV","@id":"https:\/\/geopard.tech\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/#primaryimage","url":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2025\/05\/High-Accuracy-AI-Models-Classifies-Topographic-Mapping-Faster-Than-Traditional.png?fit=1920%2C1080&ssl=1","contentUrl":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2025\/05\/High-Accuracy-AI-Models-Classifies-Topographic-Mapping-Faster-Than-Traditional.png?fit=1920%2C1080&ssl=1","width":1920,"height":1080,"caption":"High-Accuracy AI Models Classifies Topographic Mapping Faster Than Traditional"},{"@type":"BreadcrumbList","@id":"https:\/\/geopard.tech\/blog\/high-accuracy-ai-models-classifies-topographic-mapping-faster-than-traditional\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/geopard.tech\/"},{"@type":"ListItem","position":2,"name":"High-Accuracy AI Models Classifies Topographic Mapping Faster Than Traditional"}]},{"@type":"WebSite","@id":"https:\/\/geopard.tech\/#website","url":"https:\/\/geopard.tech\/","name":"GeoPard - Prec\u012bzijas lauksaimniec\u012bbas programmat\u016bra","description":"Prec\u012bz\u0101s lauksaimniec\u012bbas kart\u0113\u0161anas programmat\u016bra","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":"lv-LV"},{"@type":"Organization","@id":"https:\/\/geopard.tech\/#organization","name":"GeoPard Lauksaimniec\u012bba","alternateName":"GeoPard","url":"https:\/\/geopard.tech\/","logo":{"@type":"ImageObject","inLanguage":"lv-LV","@id":"https:\/\/geopard.tech\/#\/schema\/logo\/image\/","url":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2026\/07\/geopard_logo_transparentbackground.png?fit=512%2C68&ssl=1","contentUrl":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2026\/07\/geopard_logo_transparentbackground.png?fit=512%2C68&ssl=1","width":512,"height":68,"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\/","https:\/\/www.youtube.com\/channel\/UCiaPGLAhRPNh-s85dXdC-Sw","https:\/\/www.g2.com\/products\/geopard-agriculture-precision-farming-software\/reviews"]},{"@type":"Person","@id":"https:\/\/geopard.tech\/#\/schema\/person\/dd217733c742620adc57befbbcd84a8a","name":"dementievgeopard","image":{"@type":"ImageObject","inLanguage":"lv-LV","@id":"https:\/\/secure.gravatar.com\/avatar\/7c09adac9c20b00454199df5a41ea67c812a6c97e948bf109899836ddde31354?s=96&d=identicon&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/7c09adac9c20b00454199df5a41ea67c812a6c97e948bf109899836ddde31354?s=96&d=identicon&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/7c09adac9c20b00454199df5a41ea67c812a6c97e948bf109899836ddde31354?s=96&d=identicon&r=g","caption":"dementievgeopard"}}]}},"jetpack_publicize_connections":[],"jetpack_likes_enabled":true,"jetpack_sharing_enabled":true,"jetpack_shortlink":"https:\/\/wp.me\/pdiCPa-31o","jetpack_featured_media_url":"https:\/\/i0.wp.com\/geopard.tech\/wp-content\/uploads\/2025\/05\/High-Accuracy-AI-Models-Classifies-Topographic-Mapping-Faster-Than-Traditional.png?fit=1920%2C1080&ssl=1","_links":{"self":[{"href":"https:\/\/geopard.tech\/lv\/wp-json\/wp\/v2\/posts\/11618","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/geopard.tech\/lv\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/geopard.tech\/lv\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/geopard.tech\/lv\/wp-json\/wp\/v2\/users\/210157960"}],"replies":[{"embeddable":true,"href":"https:\/\/geopard.tech\/lv\/wp-json\/wp\/v2\/comments?post=11618"}],"version-history":[{"count":0,"href":"https:\/\/geopard.tech\/lv\/wp-json\/wp\/v2\/posts\/11618\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/geopard.tech\/lv\/wp-json\/wp\/v2\/media\/11626"}],"wp:attachment":[{"href":"https:\/\/geopard.tech\/lv\/wp-json\/wp\/v2\/media?parent=11618"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/geopard.tech\/lv\/wp-json\/wp\/v2\/categories?post=11618"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/geopard.tech\/lv\/wp-json\/wp\/v2\/tags?post=11618"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}