随机森林分类的荆门市冬闲田提取

Extraction of winter leisure fields in Jingmen City based on random forest

针对近年来冬闲田分布逐渐扩张的问题,该文提出了一种结合地表温度的冬闲田遥感提取方法。该文以Landsat 8遥感影像为数据源,利用随机森林和支持向量机的分类方法,结合地表温度因子,进行了湖北省荆门市2017年的冬闲田分布的调查。结果表明:荆门市各区县中心位置和山地区域冬闲田分布较少,丘陵梯田分布的地区冬闲田面积较大;同时考虑地表温度因子的情况下,相比于支持向量机提取方法,利用随机森林分类方法提取冬闲田取得了更好的精度。该研究说明了随机森林结合温度因素提取冬闲田可以获得较好的效果,是一种可行的冬闲田提取方法,可为合理开发利用冬闲田提供数据支撑。

Aiming at the problem of the gradual expansion of the winter leisure field in recent years,this paper proposes a remote sensing method for winter idle field combined with surface temperature. This paper uses Landsat 8 remote sensing image as the data source,and uses the classification method of random forest and support vector machine,combined with surface temperature factor,to investigate the winter field distribution in Jingmen City,Hubei Province. The results show that the central location of the districts and counties in Jingmen City and the winter area of the mountainous area are less distributed,and the area of the winter idle fields in the hilly terraces is larger. When considering the surface temperature factor,the random forest is used compared to the support vector machine extraction method. The classification method extracted winter idle fields to achieve better precision. This study demonstrates that random forest combined with temperature factors can obtain good results in winter field,and it is a feasible winter leisure field extraction method,which can provide data support for rational development and utilization of winter leisure fields.