一种反演积雪面积的范数最小二乘算法

针对常规光谱混合分析算法在积雪面积反演中存在的端元变化误差及运算效率的问题,提出了一种范数最小二乘算法(norm least squares,NLS)。为验证算法的精度和实用性,利用藏南地区的MODIS影像进行反演实验,同时采用全约束最小二乘法(fully constrained least squares,FCLS)和多端元光谱混合分析法(multiple endmember spectral mixture analysis,MESMA)进行对比分析。实验结果表明,引入范数减弱了积雪光谱的异质性,提高了积雪面积的反演精度,其反演结果基本跟真值保持一致,且用于反演积雪面积可行性高。同时,该算法反演的积雪面积相比FCLS具有更高的精度,相比MESMA具有更高的时间效率。

A Norm Least Squares Algorithm for Retrieving Snow Cover Area

Aiming at the problem of endmember variability error and computational efficiency in snow area inversion of conventional spectral hybrid analysis algorithm,a norm least squares algorithm(NLS)is proposed.In order to verify the accuracy and practicability of the algorithm,MODIS images in southern Tibet are used for inversion test,and full constrained least squares(FCLS)and multiple endmember spectral mixture analysis(MESMA)are used for comparative analysis.The experimental results show that the introduction of the norm reduces the heterogeneity of the snow spectral and improves the inversion accuracy of the snow area.The inversion results are basically consistent with the true value,and the feasibility of inverting the snow area is high.At the same time,the snow area inversion by this algorithm has higher precision than FCLS,and has higher time efficiency than MESMA algorithm.