隨著人類社會的進步以及科技快速發展下，都市版圖持續擴大，隨之衍伸出的都市空氣污染，人口密度過高，交通壅擠，等等對應問題，本研究擬以高光譜拍攝影像分辨道路與建築物之間的空間位置，透過大數據的分類技術，運用能夠增加便利與方便性也能夠改善大型運輸的交通壅擠，以及計算都市綠色植物的比例和面積以大幅改善都市的美化以及空氣汙染等問題。本研究使用資料探勘三種方法：邏輯式回歸分析法(Logistic regression）、決策樹分析演算法（Decision tree）、線性判別分析法(Linear Discriminant Analysis) 建立的三種模型並以測試範例加以分析誤差矩陣，並建構主題圖與可比對的影像差分圖，呈現本研究的成果。

With the progress of human society and the rapid development of science and technology, the scope of urban are going to be expanded. The urban air pollution, population density is too high, traffic congestion are excessively getting strong. The spatial location between roads and buildings, through the use of big data classification technology, can greatly increase the convenience of analyzing the feasibility. On the other hand, it can also improve the traffic congestion of large-scale transportation. The calculation of the proportion and area of urban green plants is to greatly improve the urban beautification and air pollution and other issues. Therefore, the study applied three different Data Mining approaches: Logistic regression, Decision Tree, and Linear Discriminant Analysis to select the required training paradigms and test paradigms to build the three models. The construction of the contrast image differences and error matrix are presented in this study.