PERANCANGAN SISTEM MONITORING CLOUD COVER UNTUK PEMANTAUAN DAN PREDIKSI CLOUD COVER MENGGUNAKAN METODE DATABASE MANAGEMENT SYSTEM DAN LONG SHORT-TERM MEMORY

Abstract

The quality of optical satellite image data obtained by the Center for Remote Sensing Data and Technology is affected by weather conditions and cloud cover. Based on these conditions, the satellite image data obtained are divided into three categories including very cloudy, cloudy, and cloud-free. Based on annual data information, it is found that the amount of cloudy satellite image data is three times greater than the amount of cloud-free satellite imagery data. So we need a system that can monitor the percentage of the extent of cloud cover from the acquisition of satellite image data. In addition, it is hoped that the creation of a system that can predict cloud cover, where the results of this cloud cover prediction can be used as a reference at the time of the next satellite image acquisition. . Through research and development of this cloud cover monitoring system, both the user and the acquisition officer can monitor the cloud cover of the acquisition result and also determine the location of cloud-free image data acquisition with predictive data. The method used for the development of the monitoring system uses a DBMS (Database Management System), while predictive research on cloud cover in an area wear the LSTM (Long short-term memory) method for Time Series Forecasting. The results of this research and development are in the form of a monitoring system that can monitor the results of acquisitions with data management principles and predict cloud cover conditions from cloud cover monitoring data.