用遥感影像提取大别山区水稻种植面积——以Landsat 8为例

本研究选取安徽省安庆市大别山为研究区,包括太湖县、岳西县、宿松县和潜山县,选取2016年和2017年的6景Landsat 8卫星遥感影像,通过遥感影像提取水稻种植面积,分析大别山区的水稻种植面积分布,并对研究区域的水稻种植面积进行动态实时监测。用遥感解译方法分别提取了研究区内的晚稻种植面积,并利用随机点验证和Kappa系数验证结果精度。结果表明:2017年和2016年水稻种植面积的提取精度分别为93.44%、93.78%,Kappa系数分别为0.86、0.83,证明水稻提取效果精确;对比研究区域各个县内的2017年和2016年晚稻种植面积,发现安庆市大别山区太湖县、潜山县、岳西县的水稻种植面积变化率均在5%以内,属于正常变化,由于2016年宿松县遭遇了水灾,部分农田被淹没,故宿松县2017年水稻种植面积相比2016年增加了13.01%;对比传统的农作物种植面积统计方法,利用遥感的方法更省人力、物力、财力,并且能精确、快速地实现对农作物的实时动态监测。

Extraction of Rice Planting Area in Dabie Mountain by Remote Sensing Image: Taking Landsat 8 as an Example

We selected Dabie Mountain in Anqing of Anhui Province as the study area, including Taihu, Yuexi, Susong and Qianshan, and analyzed the rice planting area distribution, and conducted the dynamic realtime monitoring of rice planting area by 6 Landsat 8 satellite remote sensing images from 2016 and 2017. We extracted the planting area of late rice in the study area by remote sensing interpretation method, and verified the accuracy by random point test and Kappa coefficient. The results showed that: the extraction accuracy of rice planting area in 2017 and 2016 was 93.44%, 93.78%, respectively, and the Kappa coefficient was 0.86, 0.83, respectively, proving that the extraction result was accurate; we compared the planting area of late rice in 2017 and 2016, found that the change rate of rice planting area in Taihu, Qianshan and Yuexi was less than 5%, which was a normal one; since some farmland in Susong was flooded in 2016, so its rice planting area in 2017 increased by 13.01% . Compared with traditional statistical methods of crop acreage, the method of remote sensing can save manpower, material and financial resources, and can achieve real-time and dynamic monitoring of crops accurately and quickly.