SVM算法支持下的耕地面积退化遥感监测——以昆明市呈贡区为例

Remote Sensing Monitoring of Cultivated Land Degradation Supported by SVM Algorithm—A Case of Chenggong District,Kunmin City

基于Landsat卫星TM、ETM+和OLI数据,采用SVM算法结合RS和GIS技术,分别提取了呈贡区5个时期(1989、1996、2003、2009、2015年)的耕地面积。结果表明,1989—2015年呈贡区耕地面积由156.99 km2减少到40.02 km2,减少了74.51%,年平均减少4.50 km2;空间分布上,耕地集中在呈贡西部、西南部、东北部和中部地区,其中以中部地区减少最为剧烈和明显。原因主要是政府的宏观调控、城市化速率和城市建设占用了大量耕地。建议政府协调人、地、自然和城市发展建设之间的关系,严守耕地红线,出台耕地保护措施。

It was used to extract the cultivated land area of Chenggong District that SVM algorithm combined with RS and GIS technology based on the Landsat TM, ETM ＋and OLI data in five periods （1989, 1996, 2003, 2009, 2015） respectively. The results showed that the degradation of cultivated land in Chenggong District reduced from 156.99 km^2 to 40.02 km^2 during 1989 - 2015, showed a decrease of 74. 15% and an annual reduction of 4.50 km^2 in the past 26 years. Cultivated land concentrated in the western, southwest, northeast and cen- tral parts of Chenggong in spatial distribution, of which the central region reduced the most dramatic and obvious. The main reason was the government＇ s macro-control, urbanization rate and urban construction occupied a lot of cultivated land. It was suggested that the govemment should coordinate the relationship between human, land, nature and urban development and construction, safeguard the red line of cultivated land and introduce the protection measures of cultivated land.