南方丘陵地区复杂地表“同物异谱”分类处理模型

发布时间：2018-02-28 09:31

本文关键词： 混杂地物同物异谱决策树分类南方丘陵地区　出处：《国土资源遥感》2016年02期 　论文类型：期刊论文

【摘要】：复杂流域地物混杂,因受背景干扰,对混杂在不同背景地物中的目标地物很难用1个规则区分出来。基于Landsat8多波段影像,以湘江的Ⅰ级支流蒸水河流域和长江三角洲的3大城市为研究区,采用最大似然法进行初步分类;分析混杂地物的光谱特征,探讨区分混杂地物的决策规则,对水体、人工建筑、耕地、裸地、林地和裸岩等基本地物进行分类。结果表明,蒸水河流域分类的整体精度达到88.21%,高于监督分类精度(79.68%);其他3个长江三角洲城市的分类精度都在92%以上。所提出的混杂地物区分模型可以提高混杂在不同背景中的同类地物的分类精度。
[Abstract]:Because of the background interference, it is difficult to distinguish the target objects mixed in different background objects by one rule. Based on Landsat8 multi-band images, Taking the first grade tributary of Xiangjiang River as the study area and the three major cities of the Yangtze River Delta as the study areas, the maximum likelihood method is used to carry out the preliminary classification, and the spectral characteristics of the hybrid objects are analyzed, and the decision rules for distinguishing the mixed objects are discussed. Artificial buildings, cultivated land, bare land, woodland and bare rock are classified. The results show that, The overall classification accuracy of the Yunshui River Basin reaches 88.21, which is higher than that of the supervised classification accuracy of 79.68%, and the classification accuracy of the other three cities in the Yangtze River Delta is above 92%. The proposed hybrid ground feature classification model can be improved to be mixed in different backgrounds. The classification accuracy of the same kind of features.
【作者单位】：河海大学地学院地理信息科学系;
【基金】：国家自然科学重点基金项目“半湿润半干旱流域水文模型研究”(编号:41130639)资助
【分类号】：TP751

【相似文献】