INDEKS BUAH DAN BUNGA SEBAGAI ALGORITMA DALAM MENDETEKSI DAN MENGKUANTIFIKASI FENOLOGI BUNGA DAN BUAH TANAMAN HUTAN MENGGUNAKAN CITRA UAV DAN SATELIT, SERTA HUBUNGANNYA TERHADAP PENGARUH IKLIM

• Main Author: Simarmata, Soniartha
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/50707
• Institution: Institut Teknologi Bandung

Flowering and fruiting phenology of forest species related in time and intensity of flowers and fruits which are very sensitive to climate change. So far, observations of the phenology of flowers and fruit of forest plants have been carried out using conventional methods, so it is very time consuming, labor intensive and resource intensive. Observations using remote sensing and geographic information systems have started at this time. Observation using UAV (Ummanned Aerial Vehicle) and satellite imagery provides a way out of this challenge. UAVs provide very high resolution viewing of flowers and fruit, while satellite images provide time series data. In addition, climatic data were used to see the climatic effects with flowering and fruiting. So that the purpose of this study is to measure the accuracy of the UAV index with the SVM (Support Vector Machine) classification in the monitoring of flowers and fruit, quantify the flowers and fruit results from the classification using proportion analysis in the canopy, determine the UAV index with Sentinel 1 and 2 imagery, and determine the relationship between the phenology of satellite imagery and climate. In this study, the UAV used had a resolution of 1.8 cm in making the Soil Signature and Brightness Index, while the Sentinel Image had a resolution of 10 m on RGB + NIR and polarization using SAR (Syntetic Aperture Radar), namely VV (Vertical transmit / Vertical receive). ) and VH (Vertical transmit / Horizontal receive). The climate data used in this study are temperature, humidity, rainfall, long exposure, and wind speed. In the confusion matrix analysis, the accuracy of the index accuracy of the Cassia spectabilis, Spathodea campanulata and Swietenia macrophylla flowers was 100%, 83%, 88%, while the phenological quantification obtained the proportion of flowers and fruit, namely 4.1%, 6.0%, and 2. 1%. The correlation between UAV and Sentinel using Swietenia macrophylla stands with a gap of R2 of 0.38 using VV / VH polarization. Phenological correlation of Swietenia macrophylla using VV / VH polarization with R2 of -0.81, which has an inverse significant effect (p-value <0.05) on wind speed.