基于AlexNet和支持向量机相结合的卫星遥感影像土地利用变化检测

Land Utilization Change Detection of Satellite Remote Sensing Image Based on AlexNet and Support Vector Machine

卫星遥感技术的快速发展为土地利用变化的检测提供了重要的技术支撑。为了进一步提高土地利用变化的检测精度,提出了AlexNet和支持向量机(SVM)相结合的土地利用变化分类方法。利用2013—2017年江西省南昌市的高分一号卫星遥感影像,生成该地区在这5年内的土地利用变化图,分析土地利用变化的特征。结果表明:研究区的土地类型主要以植被、水体、裸地和建筑用地为主;在这5年中,植被面积变化得最大,减少了54.74 km^2,水体面积增加了22.12 km^2,建筑用地面积增加了19.45 km^2,裸地面积增加了5.17 km^2。

The rapid development of the satellite remote sensing technology provides important technical support for land utilization change detection.To further improve the accuracy of land utilization change detection,this paper proposes a land utilization change detection method combining AlexNet and support vector machine(SVM).This method involves the use of the GF-1 satellite remote sensing images in Nanchang City,Jiangxi Province,China,from 2013 to 2017 in order to generate the land utilization change map of the area in the five years.In addition,an analysis of the land utilization change characteristics is also conducted.The results reveal that the land types in the study area are mainly vegetation,water,bare land,and building.In the past five years,the vegetation area has changed the most,which decreased by 54.74 km^2;the water area has increased by 22.12 km^2,the building area has increased by 19.45 km^2,and the bare land area has decreased by 5.17 km^2.