|
Publication
|
Title of Article |
UAV-Based Multispectral Imagery for Estimating Cassava Tuber Yields |
Date of Acceptance |
6 February 2022 |
Journal |
Title of Journal |
Engineering in Agriculture, Environment and Food (EAEF) |
Standard |
SCOPUS |
Institute of Journal |
Takahisa Nishizu |
ISBN/ISSN |
|
Volume |
|
Issue |
|
Month |
|
Year of Publication |
2022 |
Page |
|
Abstract |
This experiment aims to study the feasibility of tuber yield prediction in cassava fields using imagery acquired from a multispectral camera mounted on an Unmanned Arial Vehicle (UAV). The UAV equipped with a multispectral camera was used to fly over a cassava field once a month, from age four to seven months old (April-July). There is a total of 4 flight missions. The acquired images were generated in Pix4Dmapper to produce reflectance maps in 5 bands, Blue, Green, Red, NIR, and Red Edge band, as well as a digital surface model (DSM). The DSM map was used to calculate the height of the cassava canopy, and three reflectance maps in Red, NIR, and Red Edge band were used to calculate vegetation indices, normalized difference vegetation index (NDVI), simple ratio vegetation index (RVI), and chlorophyll vegetation index (CIrededge). Simple linear regression (SLR) developed cassava yield prediction models using NDVI, RVI, CIrededge, canopy area, and average height values calculated from 3 different methods: excluded soil pixels, zero soil pixels, and included soil pixels method, respectively. The results show that the average height and canopy area value calculated from excluded soil pixels method provides the highest R2, 0.87 and 0.65, respectively. While vegetation indices values calculated by included soil pixels method gives R2 0.58 for NDVI, R2 0.57 for CIrededge, and R2 0.50 for RVI. However, NDVI should not include any further study due to its saturation phenomenon. |
Keyword |
cassava, yield, UAV, multispectral |
Author |
|
Reviewing Status |
มีผู้ประเมินอิสระ |
Status |
ได้รับการตอบรับให้ตีพิมพ์ |
Level of Publication |
นานาชาติ |
citation |
false |
Part of thesis |
true |
Attach file |
|
Citation |
0
|
|
|
|
|
|
|