王枭轩, 左小清, 孟庆岩, 占玉林, 刘苗, 杨泽楠, 李雨秦. 基于高分3号和高分2号叶面积指数反演与分析[J]. 云南大学学报(自然科学版), 2019, 41(4): 731-737. doi: 10.7540/j.ynu.20180678
引用本文: 王枭轩, 左小清, 孟庆岩, 占玉林, 刘苗, 杨泽楠, 李雨秦. 基于高分3号和高分2号叶面积指数反演与分析[J]. 云南大学学报(自然科学版), 2019, 41(4): 731-737. doi: 10.7540/j.ynu.20180678
WANG Xiao-xuan, ZUO Xiao-qing, MENG Qing-yan, ZHAN Yu-lin, LIU Miao, YANG Ze-nan, LI Yu-qin. Inversion and analysis of leaf area index based on GF-3 and GF-2[J]. Journal of Yunnan University: Natural Sciences Edition, 2019, 41(4): 731-737. DOI: 10.7540/j.ynu.20180678
Citation: WANG Xiao-xuan, ZUO Xiao-qing, MENG Qing-yan, ZHAN Yu-lin, LIU Miao, YANG Ze-nan, LI Yu-qin. Inversion and analysis of leaf area index based on GF-3 and GF-2[J]. Journal of Yunnan University: Natural Sciences Edition, 2019, 41(4): 731-737. DOI: 10.7540/j.ynu.20180678

基于高分3号和高分2号叶面积指数反演与分析

Inversion and analysis of leaf area index based on GF-3 and GF-2

  • 摘要: 针对光学数据反演叶面积指数(LAI)容易受到云雾遮挡和光学遥感信息饱和的问题,根据雷达散射机制和Yamaguchi分解,提出了极化分解植被指数,利用光学植被指数和极分解植被指数融合形成光学与微波极化分解融合植被指数;利用光学与微波极化分解融合植被指数与实测数据建立回归模型反演叶面积指数,并对该模型精度评价. 实验表明:光学与微波极化分解融合植被指数与实测数据建立回归模型反演叶面积指数的精度要优于极化分解植被指数和光学植被指数与实测数据建立的回归模型,其中MRVI与LAI建立回归模型是最优的,R2达到0.7262,RMSE到达0.3548. 综上所述,光学与微波极化分解融合植被指数不仅充分利用雷达能够穿透浓密植物的特性,而且融合光学数据对叶面积指数的反演敏感性,更能准确的反演叶面积指数.

     

    Abstract: The inversion of leaf area index (LAI) for optical data is susceptible to cloud occlusion and saturation of optical remote sensing information. Based on the radar scattering mechanism and Yamaguchi decomposition, this paper proposes a polarization decomposition vegetation index, and uses the optical vegetation index and the polar decomposition vegetation index to form an optical and microwave polarization decomposition fusion vegetation index, which, combined with the measured data, is used to establish a regression model so as to invert the leaf area index, and evaluate the accuracy of the model. Experiments show that the optical and microwave polarization decomposition fusion vegetation index and measured data establish a regression model that inverts the accuracy of leaf area index better than does the regression model established by polarization decomposition vegetation index and optical vegetation index and measured data, among which the regression model formed by MRVI and LAI is optimal, with R2 reaching 0.726 2 and RMSE reaching 0.354 8. In summary, the optical and microwave polarization decomposition fusion vegetation index are able to fully utilize the characteristics of radar capable of penetrating dense plants; the inversion sensitivity of the fusion optical data to the leaf area index is capable of inverting the leaf area index more accurately.

     

/

返回文章
返回