CAJ | 학술논문

以云南省文山壮族苗族自治州麻栗坡县2005年T M影像为试验数据，利用最大似然分类（M LC ）、支持向量机（SVM ）以及随机森林（RF）3种分类方法进行了土地覆盖遥感分类研究。从分类精度、样本数量对分类器的影响、模型复杂度、分类速度等几个方面进行了比较分析。结果表明：随机森林分类法最优，而经典方法之一的最大似然分类法最稳定。所得出的结论对在类似的应用中如何选择合适的分类方法具有一定的参考价值。
이운남성문산장족묘족자치주마률파현2005년T M영상위시험수거，이용최대사연분류（M LC ）、지지향량궤（SVM ）이급수궤삼림（RF）3충분류방법진행료토지복개요감분류연구。종분류정도、양본수량대분류기적영향、모형복잡도、분류속도등궤개방면진행료비교분석。결과표명：수궤삼림분류법최우，이경전방법지일적최대사연분류법최은정。소득출적결론대재유사적응용중여하선택합괄적분류방법구유일정적삼고개치。
This article usesMaximum Likelihood Classification (MLC) ,Support Vector Machine (SVM) and Random Forest (RF) to study the land cover classification based on the Thematic Mapper (TM ) images of 2005 in Malipo County ,Wenshan Zhuang Prefecture in Yunnan Province .And then it carries out a compara-tive analysis of the classification results of three classifiers from the aspects of classification accuracy ,model complexity ,and time efficiency .The results show that RF is the best and MLC is more stable than other two methods .Therefore ,the conclusions in this study are valuable for how to select classifiers in the similar ap-plications .