本文選題：NDVI遙感數(shù)據(jù) + 植被退化趨勢�。� 參考：《科技通報》2016年11期

【摘要】：為了分析貴州烏蒙山綠色植被退化趨勢,提出針對貴州烏蒙山區(qū)綠色植被覆蓋率退化趨勢的圖像分析方法。獲取NDVI遙感圖像數(shù)據(jù)信息,運用基于紋理特征的ISODATA算法提取植被圖像中每個像素點的紋理特征、位置特征等綜合特點,形成特征矢量結(jié)構(gòu)空間;在特征空間中,利用ISODATA算法主動調(diào)整參數(shù)獲取初始聚類數(shù)量及聚類中心后,進行區(qū)域分割,獲取圖像區(qū)域特征信息;然后建立像元二分模型對植被覆蓋度圖像特征信息進行分析,通過計算不同年份植被覆蓋率對綠色植被覆蓋率退化趨勢進行預測。實驗結(jié)果證明,改進的圖像分析方法可以對貴州烏蒙山區(qū)綠色植被覆蓋率退化趨勢進行準確分析,精度較高。
[Abstract]:In order to analyze the trend of green vegetation degradation in Wumeng Mountain, Guizhou Province, an image analysis method for the degradation trend of green vegetation coverage in Wumeng Mountain area of Guizhou Province was proposed. The NDVI remote sensing image data information is obtained, and the texture feature and position feature of each pixel in vegetation image are extracted by ISODATA algorithm based on texture feature to form the feature vector structure space. ISODATA algorithm is used to adjust the parameters to get the initial cluster number and cluster center, then the region is segmented to obtain the regional feature information of the image, and then the pixel dichotomy model is established to analyze the feature information of vegetation coverage image. The degradation trend of green vegetation coverage was predicted by calculating vegetation coverage in different years. The experimental results show that the improved image analysis method can accurately analyze the degradation trend of green vegetation coverage in Wumeng Mountain area of Guizhou Province with high accuracy.
【作者單位】： 貴州商學院計算機與信息工程學院;貴州大學計算機科學與技術(shù)學院;
【分類號】：TP751
，

本文編號：1831073