水保局和農委會共同推廣「健康、效率、永續經營」農業施政方針， 特別鎖定了高鐵沿線左右各1.5 公里範圍內的農業用地，將對此地區進行全 面性的農業輔導，因此又稱為黃金廊道，此區域稻米豐富，而水稻對綠環境 影響甚大。高光譜是一種先進的影像材料， 這種材料資訊豐富的影像在判 釋上亦需要花費較多的時間，若能建構影像決策系統進行分析，準確的判讀 出地表上所代表的物種，就能大幅減少實地探勘的人力、物力與時間。本研 究主要探討如何從高光譜影像中篩選出重要的光譜資訊，並以水稻田為主要 判釋對象，搭配監督式類神經網路(Back-propagation neural network; BPN)的 分類器進行判釋。 設計以下兩種平行研究案例: (a)原始波段搭配倒傳遞類神 經網路 (b)亂度基礎分類法(entropy-based classification; EBC)篩選出重要光 譜資訊搭配倒傳遞類神經網路。研究分析出高光譜和多光譜的判釋差異和主 題圖的比較。透過高光譜原始資料於倒傳遞類神經網路之正確率約為99.4%， 經過屬性刪減後正確率為96.35%，但主題圖較完整， 椒鹽效應比原始資料 經倒傳遞類神經網路分類較少。

The Soil Water Conservation Bureau and Council of Agriculture has promote the “Healthy, efficiency, and Sustainable development” for target policy. The range of high-speed railway with 1.5 kilometer is the agriculture used area for guidance which is called “golden gallery”. In this area, the paddy rice is abundant and quite influence for environments. The hyperspectral is an advanced material which renders more information for image classification. However, spectral information is very rich which induced the computation also takes more time. Therefore, if we can construct a classification system to accurately interpreting the surface of ground, it can be applied significantly to reduce manpower and time. This study focused on how to extract the important factors of hyperspectral image to classify the paddy rice area with applying supervised Back propagation Neural Network algorithm. There are two parallel studies designed as (a) the original band with a back-propagation neural network (b) entropy-base-classification filter out important information with back-propagation neural network. The accuracy rate and thematic map are shown.