论文部分内容阅读
以呼和浩特市土默特左旗白石头沟林场为研究对象,利用ENVI4.7对研究区TM数据进行处理,建立基于马氏距离分类的主要树种的训练样本,并提取训练样本TM4、3、2波段的灰度值作为特征向量,取其均值代入基于两个总体的马氏距离判别公式,建立判别函数判定未知类别。结合研究区的森林资源二类调查数据,进行分类结果检验,得到总体分类精度为67.73%,落叶松林、油松林、白桦林的分类精度分别为39.79%、41.62%、88.17%。
Taking Baishitougou Forest Farm of Tumotezuoqi, Hohhot as the research object, the TM data of the study area were processed by ENVI4.7, and the training samples of the main tree species based on Mahalanobis distance were established, and the training samples TM4,3,2 Band gray value as the eigenvector, and take the average value into the Mahalanobis distance discriminant formula based on two populations to establish the discriminant function to determine the unknown category. According to the second category of forest resources survey data in the study area, the classification results were tested and the overall classification accuracy was 67.73%. The classification accuracies of Larix gmelinii, Pinus tabulaeformis and Betula platyphylla were 39.79%, 41.62% and 88.17% respectively.