| 摘要: |
| 【目的】获取内蒙古沿黄平原区农业资源信息，提高大区域中低分辨率遥感影像上作物种植结构分类的效率和精度。【方法】多时相作物分类方法充分利用作物的季候特征，能够清晰地反映不同作物随时间的变化趋势，能够有效地减少作物误分类现象。在内蒙古沿黄平原区构建增强型植被指数（EVI）时间序列曲线，依据不同农作物EVI时间序列曲线的差异对6种主要农作物小麦、玉米、葵花、西葫芦、番茄和苜蓿的种植结构进行了分析。【结果】小麦、玉米、葵花、西葫芦、番茄、苜蓿和其他作物的用户精度分别为：79.59%、80%、83.67%、78.18%、75.93%、82.22%、68.75%，制图精度分别为78%、80%、82%、86%、82%、74%、66%，农作物总体分类精度达到78.29%，kappa系数为0.747。经统计沿黄灌区种植的玉米实地统计面积合计为7,912.17 km2,文中提取出的面积为7,412.75 km2，相对误差为6.31%。【结论】通过对MODIS-EVI时间序列的分析可以较为准确地识别大尺度测区内的主要作物，该方法能够在大区域中低分辨率影像上实现较好的分类结果。 |
| 关键词: MODIS；EVI；时间序列；种植结构 |
| DOI： |
| 分类号:TP79 |
| 基金项目:国家自然科学基金项目；内蒙古自治区科技计划项目 |
|
| Analysis of crop planting structure in the the plain along the Yellow River of Inner Mongolia Based on MODIS-EVI time series |
|
jiabozhong, baiyanying, weizhanmin
|
|
inner mongolia agricultural university
|
| Abstract: |
| 【Background】Crop planting structure is the main basis for crop growth monitoring and yield estimation analysis, crop planting structure adjustment and optimization, and crop irrigation management.Compared with the time-consuming and laborious traditional manual investigation, remote sensing technology has the advantages of low cost, short observation period and wide coverage.It is widely used in the extraction of crop planting structure, and gradually become a popular method.Therefore, how to apply remote sensing images and adopt appropriate classification methods to obtain higher accuracy classification results has become an important direction of low-cost and high-precision agricultural remote sensing research.Remote sensing extraction of crop planting structure is the main method of crop planting structure extraction.【Objective】To obtain agricultural resource information along the Yellow Plain in Inner Mongolia, and improve the efficiency and accuracy of crop planting structure classification in medium and low resolution remote sensing image in large area.【Method】Multi-temporal crop classification method makes full use of crop seasonal characteristics, can clearly reflect the change trend of different crops over time, and can effectively reduce crop misclassification phenomenon.A time series curve of enhanced vegetation index (EVI) was constructed along the Yellow Plain in Inner Mongolia. According to the difference of EVI time series curve of different crops, the planting structure of wheat, corn, sunflower, zucchini, tomato and alfalfa was analyzed.【Result】The user accuracy of wheat, corn, sunflower, zucchini, tomato, alfalfa and other crops is 79.59%, 80%, 83.67%, 78.18%, 75.93%, 82.22% and 68.75% respectively, and the mapping accuracy is 78%, 80%, 82%, 86%, 82%, 74%, 74% and 66%, respectively. The overall classification accuracy of crops is 78.29%, and the kappa coefficient is 0.747.According to statistics, the total field statistical area of maize planted along the Yellow River irrigation area is 7,912.17 km2, and the area extracted in this paper is 7,412.75 km2, and the relative error can be obtained as 6.31%.【Conclusion】By analyzing the time series based on MODIS-EVI, the main crops in the large-scale survey area can be identified more accurately, and the method can achieve better classification results in the low-resolution images in the large area. |
| Key words: MODIS;EVI;Time series;Planting structure;the plain along the Yellow River |
