| 摘要: |
| 【目的】有效、精确地获取乡镇尺度农作物种植面积信息。【方法】利用高精度遥感影像信息，采用最大似然法进行地物分类，对研究区主要农作物种植面积进行了估算，并通过地面抽样调查信息作为辅助参量进行了精度评价。【结果】采用高精度遥感图像结合最大似然法可以有效、精确地提取乡镇尺度农作物种植面积，玉米和葵花的用户精度均高于91%，在遥感影像受天气影响较小的2013年和2015年，各用户精度达到95%左右，Kappa系数达到0.94。研究区玉米的种植面积逐年扩大，而葵花的种植面积逐年缩小。同时由于农田改造，使得非种植区的面积有一定程度的减少。【结论】高分辨率卫星图像能帮助农民和政府低成本地估算农作物种植面积，进而调整耕地，优化种植结构。 |
| 关键词: 遥感； 监测； 种植面积；尺度； 灌区 |
| DOI：10.13522/j.cnki.ggps.2017.0268 |
| 分类号: |
| 基金项目: |
|
| Monitoring Cropland Types at Village-town Scale in Hetao Irrigation District Using High-resolution Satellite Images |
|
CHEN Zhisen， SI Bingcheng
|
|
Key Laboratory of Agricultural Soil and Water Engineering in Arid Area, Ministry of Education, Northwest A&F University, Yangling 712100, China; 2.Institute of Water-Saving Agriculture in Arid Area of China, Northwest A&F University, Yangling 712100, China
|
| Abstract: |
| 【Objective】 Knowing the cultivation areas of different crops in a catchment or irrigation district is essential to improve agricultural management, and the purpose of this paper is to study the feasibility of using high-resolution satellite images to monitor the crop maps at village-town scale. 【Method】 We took Hetao Irrigation District as an example, the cultivation areas of different crops in which were calculated using the maximum likelihood method based on high-resolution remote sensing images. The accuracy of the calculated crop maps was verified against field survey data. 【Result】 High-resolution remote sensing images and the maximum likelihood method were efficacy to estimate crop maps at village-town scale in this region. Compared with the survey data, the error of the cultivated areas of maize and sunflower estimated by the proposed method was less than 9%. The images in 2013 and 2015 were least impacted by weather, and the accuracy of all estimated results was approximately 95% with a Kappa of 0.94. It was found that the cultivated area of maize has been increasing at the expense of sunflower and uncultivated land due to agricultural production reconstruction. 【Conclusion】 Combining high-resolution remote sensing images and the maximum likelihood method proved to be an effective and accurate method to estimate crop maps at village-town scale. |
| Key words: remote sensing; monitoring; planting area; scale； irrigation district |
