< img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=1241806559960313&ev=PageView&noscript=1" /> Izindaba - I-UAV Multispectral Remote Sensing yokuqapha ukukhula kukakotini

I-UAV Multispectral Remote Sensing yokuqapha ukukhula kukakotini

Ukotini njengesivuno esibalulekile semboni yendwangu kanye nemboni yendwangu kakotini, ngokwanda kwezindawo ezinabantu abaningi, ukotini, okusanhlamvu nezitshalo zembewu inkinga yokuncintisana komhlaba iya ngokuya iba mbi kakhulu, ukusetshenziswa kukakotini nokutshala okusanhlamvu kunganciphisa ngempumelelo ukungqubuzana phakathi ukutshalwa kwezitshalo zikakotini nokusanhlamvu, okungathuthukisa ukukhiqizwa kwezitshalo nokuvikelwa kokuhlukahluka kwemvelo nokunye. Ngakho-ke, kubaluleke kakhulu ukuqapha ngokushesha nangokunembile ukukhula kukakotini ngaphansi kwemodi ye-intercropping.

I-UAV-Multispectral-Remote-Sensing-to-Monitor-Cotton-Growth-1

Izithombe eziningi zikakotini ezibukwayo nezibonakalayo ezigabeni ezintathu zokuzala zatholwa yizinzwa ze-UAV-mounted multi-spectral kanye ne-RGB, izici zazo ze-spectral nezithombe zakhishwa, futhi zahlanganiswa nokuphakama kwezitshalo zikakotini emhlabathini, i-SPAD kakotini yakhishwa. kulinganiselwa ngokufunda okudidiyelwe kokuhlehla kokuvota (VRE) futhi uma kuqhathaniswa namamodeli amathathu, okuyi-Random Forest Regression (RFR), Gradient Boosted Tree Regression (GBR), kanye ne-Support Vector. Ukwehla komshini (SVR). . Sihlole ukunemba kwesilinganiso samamodeli ahlukene wokulinganisa kokuqukethwe kwe-chlorophyll ehlobene kakotini, futhi sahlaziya imiphumela yezilinganiso ezihlukene zokutshala phakathi kukakotini nobhontshisi wesoya ekukhuleni kukakotini, ukuze sinikeze isisekelo sokukhethwa kwesilinganiso sokutshala ngokuhlanganyela. phakathi kukakotini nobhontshisi wesoya kanye nesilinganiso sokunemba okuphezulu sikakotini we-SPAD.

Uma kuqhathaniswa namamodeli e-RFR, GBR, kanye ne-SVR, imodeli ye-VRE ibonise imiphumela engcono kakhulu yokulinganisa ekulinganiseni i-SPAD kakotini. Ngokusekelwe kumodeli yokulinganisa ye-VRE, imodeli enezici zesithombe esibonakala ngezinto eziningi, izici zesithombe ezibonakalayo, kanye nokuhlanganiswa kobude besitshalo njengoba okokufaka kube nokunemba okuphezulu kakhulu ngesethi yokuhlola engu-R2, i-RMSE, ne-RPD engu-0.916, 1.481, ne-3.53, ngokulandelana.

I-UAV-Multispectral-Remote-Sensing-to-Monitor-Cotton-Growth-2

Kwaboniswa ukuthi ukuhlanganiswa kwedatha yemithombo eminingi kuhlanganiswe ne-algorithm yokuhlanganiswa kokuhlehla kokuvota kunikeza indlela entsha nephumelelayo yokulinganisa kwe-SPAD kukotini.


Isikhathi sokuthumela: Dec-03-2024

Shiya Umlayezo Wakho

Sicela ugcwalise izinkambu ezidingekayo.