TY - GEN
T1 - Prediction of Rice Growth Phases with Multitemporal Landsat-8 Data Using Rotation Forest Multiclass Method
AU - Novidianto, Raditya
AU - Fithriasari, Kartika
AU - Kuswanto, Heri
N1 - Publisher Copyright:
© 2023 American Institute of Physics Inc.. All rights reserved.
PY - 2023/1/27
Y1 - 2023/1/27
N2 - In 2018 the Area Sample Framework Survey (ASF) was formed, which was carried out by BPS Statistics to calculate rice harvested area and improve food crop data. The combination of satellite data and official data is an innovation that needs to be done to overcome a limitation, especially in the Sampling ASF carried out by BPS Statistics, so that the success of combining official data and big data will make suggestions for adding samples to the non-sample ASF Survey for data estimation harvest area is more accurate. Rotation Forest is a method that is often used and excels in classification with continuous data predictors. Multitemporal remote sensing using Landsat-8 satellite imagery was launched in 2013 with a recording period every 16 days. The basic features produced on the Landsat-8 satellite include bands 1 to 7, EVI, NDVI, NDWI, and NDBI indexes that can be used for prediction using the ensemble rotfor method. OVO method is better than the OVA in the case of multiclass rotfor rice growth phase detection using Landsat-8 satellite imagery. The best model formed is the RotFor MultiClass OVO model with a sensitivity value of 0.88, specificity 0.96, accuracy 0.87, MCC 0.83 and Cohen Kappa Index 0.83.
AB - In 2018 the Area Sample Framework Survey (ASF) was formed, which was carried out by BPS Statistics to calculate rice harvested area and improve food crop data. The combination of satellite data and official data is an innovation that needs to be done to overcome a limitation, especially in the Sampling ASF carried out by BPS Statistics, so that the success of combining official data and big data will make suggestions for adding samples to the non-sample ASF Survey for data estimation harvest area is more accurate. Rotation Forest is a method that is often used and excels in classification with continuous data predictors. Multitemporal remote sensing using Landsat-8 satellite imagery was launched in 2013 with a recording period every 16 days. The basic features produced on the Landsat-8 satellite include bands 1 to 7, EVI, NDVI, NDWI, and NDBI indexes that can be used for prediction using the ensemble rotfor method. OVO method is better than the OVA in the case of multiclass rotfor rice growth phase detection using Landsat-8 satellite imagery. The best model formed is the RotFor MultiClass OVO model with a sensitivity value of 0.88, specificity 0.96, accuracy 0.87, MCC 0.83 and Cohen Kappa Index 0.83.
KW - Area Sampling Framework
KW - Binarization
KW - Classification
KW - Landsat-8
KW - Rotation Forest
UR - http://www.scopus.com/inward/record.url?scp=85147307777&partnerID=8YFLogxK
U2 - 10.1063/5.0107155
DO - 10.1063/5.0107155
M3 - Conference contribution
AN - SCOPUS:85147307777
T3 - AIP Conference Proceedings
BT - 3rd International Conference on Science, Mathematics, Environment, and Education
A2 - Indriyanti, Nurma Yunita
A2 - Sari, Meida Wulan
PB - American Institute of Physics Inc.
T2 - 3rd International Conference on Science, Mathematics, Environment, and Education: Flexibility in Research and Innovation on Science, Mathematics, Environment, and Education for Sustainable Development, ICoSMEE 2021
Y2 - 27 July 2021 through 28 July 2021
ER -