测绘通报 ›› 2018, Vol. 0 ›› Issue (9): 82-86.doi: 10.13474/j.cnki.11-2246.2018.0285

• 技术交流 • 上一篇    下一篇

极限学习机辅助下路域植被叶面积指数的反演

雷宇斌1, 朱善宽2,3, 郭云开3, 李丹娜3, 刘磊3, 刘宁3   

  1. 1. 湖南省第二测绘院, 湖南 长沙 410000;
    2. 中交上海航道勘察设计院有限公司, 上海 200000;
    3. 长沙理工大学测绘遥感应用技术研究所, 湖南 长沙 410076
  • 收稿日期:2018-05-22 修回日期:2018-07-20 出版日期:2018-09-25 发布日期:2018-09-29
  • 作者简介:雷宇斌(1971-),男,高级工程师,主要研究方向为测绘及遥感理论应用研究。E-mail:2118949836@qq.com
  • 基金资助:

    国家自然科学基金面上项目(41471421;41671498)

Inversion of Leaf Area Index Based on Extreme Learning Machine Regression in Road Vegetation

LEI Yubin1, ZHU Shankuan2,3, GUO Yunkai3, LI Danna3, LIU Lei3, LIU Ning3   

  1. 1. The Second Surveying and Mapping Institute of Hunan Province, Changsha 410000, China;
    2. Shanghai Waterway Engineering Design and Consulting Company Limited, Shanghai 200000, China;
    3. Institute of Surveying and Mapping and Remote Sensing Applied Technology, Changsha University of Science & Technology, Changsha 410076, China
  • Received:2018-05-22 Revised:2018-07-20 Online:2018-09-25 Published:2018-09-29
  • Contact: 朱善宽。E-mail:2506272955@qq.com E-mail:2506272955@qq.com

摘要:

路域植被叶面积指数(LAI)的获取对于路域植被长势和健康状况的监测具有重要意义。本文以GF-1影像和地面同步实测数据为基础,利用极限学习机(ELM)对湖南省醴潭高速路域植被LAI进行了建模反演。试验结果表明,与传统经验回归模型、SVM模型相比,ELM反演精度更高,RMSE为0.501,预测精度为86.26%。该研究可为路域植被健康评估提供参考。

关键词: 遥感, 叶面积指数, 反演, GF-1影像, 极限学习机

Abstract:

The acquisition of LAI of road vegetation has an important implication for the monitoring of road vegetation growth and health status.Based on the GF-1 image data and the ground synchronous measured LAI data,the technology of extreme learning machine (ELM) was introduced to inversion modeling in the high-speed road area of Hunan Province.Compared with the traditional empirical regression and support vector machine (SVM) methods,the extreme learning machine is higher,RMSE is 0.501,and the prediction accuracy is 86.26%.The study can provide a reference for road vegetation health assessment.

Key words: remote sensing, leaf area index, inversion, GF-1 imaging, extreme learning machine

中图分类号: