Litcius/Paper detail

Deep Learning Algorithms to determine Drought prone Areas Using Remote Sensing and GIS

Muhammad Iqbal Habibie, Tofael Ahamed, Ryozo Noguchi, Shusuke Matsushita

202033 citationsDOI

Abstract

Climate change has had a global effect on staple crops. Indonesia is a developed country facing a significant threat to climate change. The study uses the Normalized Difference Water Index (NDWI) obtained from Landsat 8 OLI to define the water scarcity in the study area. This research proposes a CNN-based YOLO model that can detect Drought in growing maize development stages. The study was observed in 2018. The detection drought based on the growing season using deep learning was found IoU, Precision, Recall, F1-Score, mean Average Precision (mAP), 83.4%, 98%, 99%, 98%, 96% in the drought-prone areas. The model allows combining remote sensing technology to detect object detection in real-time with acceptable accuracy.

Topics & Concepts

Water scarcityRemote sensingClimate changeDeep learningComputer scienceEnvironmental scienceChange detectionAlgorithmArtificial intelligenceGeographyAgricultureGeologyArchaeologyOceanographyWetland Management and ConservationWater and Land ManagementData Mining and Machine Learning Applications
Deep Learning Algorithms to determine Drought prone Areas Using Remote Sensing and GIS | Litcius