监督分类和决策树方法提取钦州湾湿地的对比分析

湿地生态系统对环境和社会经济的可持续发展有着举足轻重的作用。基于2010 年的TM影像，经过缨帽变换后，运用监督分类、决策树方法提取钦州湾的滨海湿地；以人工目视判别解译的湿地面积作为判别依据。结果表明：运用同一评价模板，监督分类、决策树2 种分类方法的精度分别为92.00%，89.00%，Kappa 系数分别为0.8952、0.8582；目视解译的湿地面积为218.30 km2，2 种方法得到的湿地面积分别为219.00 km2、193.70 km2。监督分类提取钦州湾滨海湿地信息的效果比决策树好。

Contrast Analysis of Supervised Classification and Decision Tree Method to Extract the Qinzhou Bay Wetland

Wetland ecosystem played an important role in the environment and sustainable socio-economic development. Based on the TM images in 2010 with a pretreatment of Tasseled Cap transformation, two different methods are used to extract the Qinzhou Bay coastal wetlands: Supervised Classification (SC) and Decision Trees (DT). Coastal wetlands were picked out by artificial visual interpretation as discriminant standard. The result showed that when the same evaluation template was used, the accuracy and Kappa coefficient of SC, DT was 92.00%, 0.8952 and 89.00%, 0.8582 respectively. The total area of coastal wetland was 218.3 km2 by artificial visual interpretation, and the extracted wetland area of SC, DT was 219 km2 and 193.70 km2 respectively. The result indicated that, for Qinzhou bay coastal wetland information extraction, the effect of the Supervised Classification (SC) was better than the Decision Tree (DT).