2017-01-012024-05-15https://scholars.lib.ntu.edu.tw/handle/123456789/664213摘要：本研究以105年「利用UAV、航遙測資訊與多尺度空間資訊轉換分類模式強化水稻田面積及產量調查技術研究」計畫執行成果為基礎，除持續發展多源多時期遙測影像資料於水稻田面積判釋及產量推估技術發，增加以多時序遙測影像資料分析水稻之光譜反射率及雷達散射特徵，進而發展稻作坵塊之多時序遙測影像偵測模式原型，以期解決各地區因氣候條件差異導致稻作插秧時序不一致，進而產生之水稻判釋精度瓶頸。依據各項研究執行成果，本研究將彙整並評估研究成果之業務推動的執行成本與偵測效益，並提出以遙測資訊進行水稻分布監測及產量推估的執行方案建議。<br> Abstract: Based on the previous research, Research of Apply UAV, Remote Sensing Information and Multi-Scale Spatial Information Transform Classification Model to Improve Paddy Rice Area and Yield Investigation Technique, in 2016. This research will not only continue to improve upon the interpretation accuracy and efficiency by integrating optical satellite, aerial image, satellite SAR image and UAV VIS/NIR/SAR images for paddy rice distribution and production estimation, a new method of using pattern changes through optical spectral reflectance and SAR scatter features between multi-temporal remote sensing data is expected to develop a prototype of multi-temporal rice fields interpretation model to distinguish rice fields which transplanted in different weather conditions. According to the results, an evaluation will be completed for the balance between data accuracy, processing cost and interpret benefit and used as area of interest such as rice distribution monitoring, production estimation and subsequent statistical analysis.多尺度多源遙測資料水稻產量推估分類誤差分析影像時序特徵萃取Multi-Scale and Multi-Source Remote Sensing DataRice Yield PredictionClassification Error AnalysisImage Temporal Feature Extration