基于不同分类器的农用地分类提取

【目的】分析Landsat 8 OLI卫星遥感影像数据面向农用地分类的实际应用方法和效果,以新疆奇台县南部为研究对象。【方法】使用随机森林(RF)、支持向量机(SVM)和神经网络(Neural Net)三种分类器进行研究区农用地分类对比。【结果】通过对三种分类器参数设置参数精度检验,利用上述三种算法对农用地地物分类进行精度评价,在整体分类精度中,支持向量机算法(SVM)<随机森林算法(RF)<神经网络算法(Neural Net),分类精度分别为:90.75%,94.30%和94.84%。【结论】神经网络方法(Neural Net)在该地区的农用地物整体分类上,比支持向量机(SVM)和随机森林法(RF)相比具有一定的优势,并获得较好的分类精度。

Study on Classification and Extraction of Agricultural Land in Qitai County of Xinjiang Based on Different Classifiers

【Objective】 Classification based on remote sensing image is one of the important contents of remote sensing data application. How to improve the classification accuracy of remote sensing image is the key of remote sensing image research. 【Method】In order to analyze the practical application method and effect of Landsat 8 OLI satellite remote sensing image data for agricultural land classification, this paper takes the southern Qitai County of Xinjiang as the research object, and uses three classifiers, random forest (RF), support vector machine (SVM) and neural network (Neural Net), to conduct a comparative study of agricultural land classification in the study area. 【Result】Through the parameter setting accuracy test of the three classifiers, the accuracy of agricultural land classification is evaluated by using the above three algorithms. In the overall classification accuracy, the support vector machine algorithm (SVM) < random forest algorithm (RF) < neural network algorithm (Neural Net) has the classification accuracy of 90.75%, 94.30% and 94.84%, respectively. 【Conclusion】Neural Net has some advantages over Support Vector Machine (SVM) and Random Forest Method (RF) in the overall classification of agricultural land use in this area, and achieves better classification accuracy.