江苏省科技期刊可视化阅读平台

导航条

江苏科技信息 ›› 2019, Vol. 36 ›› Issue (11): 53-56.doi: 10.1004-7530/2019-36-11-53

• 基础研究 • 上一篇    下一篇

南丰县桔园空间分布信息的遥感提取

文超,陆琴   

  1. 江西师范大学 地理与环境学院,江西 南昌 330022
  • 出版日期:2019-04-20 发布日期:2019-07-09
  • 作者简介:文超（1992— ）,男,陕西汉中人,硕士研究生;研究方向：土地利用,覆被变化研究。
  • 基金资助:
    国家自然科学基金(项目编号：41440004);江西省自然科学基金(项目编号：20151BAB203041)

Remote sensing extraction of spatial distribution information of citrus garden in Nanfeng county

Chao Wen,Qin Lu   

  1. College of Geography and Environment, Jiangxi Normal University, Nanchang 330022, China
  • Online:2019-04-20 Published:2019-07-09

RichHTML

2241

PDF

1250

摘要：

柑桔是我国南方亚热带红壤地区重要的经济果类,准确地获取柑桔种植分布信息对区域水土保持、制定农业政策等方面具有重要的意义。文章以江西省南丰县为研究区,利用增强自适应时空融合模型ESTARFM,对Landsat NDVI数据与同期MODIS NDVI数据进行融合并重构出高时空分辨率数据,基于此数据采用随机森林分类方法提取桔园信息。结果显示：（1）ESTARFM时空融合模型生成的影像与同期原始影像显著相关（p<0.01）,说明重构的高时空分辨率数据十分适用于研究区桔园信息的提取。（2）研究区桔园信息提取精度为95.25％,Kappa系数为0.922 7。基于高时空分辨数据,随机森林分类方法在研究区桔园提取中取得了理想的效果,为南方地区果园信息提取提供一种新的方法。

关键词: 遥感, ESTARFM, 时空融合, 桔园

Abstract:

Citrus is an important economic fruit in the subtropical red soil region of southern China. It is of great significance to obtain the information of citrus planting and distribution accurately for regional soil and water conservation and formulation of agricultural policies. In this paper, Nanfeng county of Jiangxi province is taken as the research area, and the enhanced adaptive spatio-temporal fusion model ESTARFM, is used to fuse the Landsat NDVI data and the MODIS NDVI data at the same time and to reconstruct the high spatial-temporal resolution data. Based on this data, the information of orange orchard is extracted by random forest classification. The results show that （1） the image generated by the ESTARFM spatiotemporal fusion model is significantly correlated with the original image （p<0.01）, indicating that the reconstructed high temporal and spatial resolution data is very suitable for the extraction of citrus orchard information in the study area. （2） The accuracy of the information extraction of the citrus garden is 95.25%, and the Kappa coefficient is 0.922 7. Based on the high spatial-temporal resolution data, the random forest classification method has achieved an ideal effect in the orange orchard extraction in the study area, which provides a new method for the information extraction of orchards in southern China.

Key words: remote sensing, ESTARFM, space-time fusion, citrus garden

中图分类号: