摘要
考虑影响因素的复杂性,提出土壤有机质含量区间值高光谱估测的思想。根据在陕西省横山县采集的84个土壤样本数据,采用14种光谱反射率变换方法及因子互乘变换筛选反演因子,采用模糊识别方法进行土壤有机质含量估测。结果表明,原始光谱反射率(R)及其平方(R2)、平方根(R1/2)、倒数(1/R)、自然对数(ln R)的一阶微分、二阶微分及其互乘变换与有机质含量的相关性明显增强,模型优化系数可调节类别判别的准确度,12个检验样本的准确度为91.67%。这表明提出的土壤有机质含量区间值高光谱估测模型是有效的。
As for the complex nature of the hyper-spectral estimation problem, the thought of the interval estimation of soil organic matter content is put forward in this paper. According to hyper-spectral data of the soil spectral and soil organic matter content data at laboratory in Shanxi province Hengshan county, the inversion factors are selected by adopting the fourteen transforms and mutual multiplication, and the soil organic matter content interval is estimated by using fuzzy recognition method. The experiment results show that the correlation between the original spectral reflectance( R), square( R^2), square root( R^1/2), reciprocal( 1 / R), natural logarithm( ln R) and soil organic matter content is less, but it is enhanced obviously for their first derivative and second derivative, and for the product of original inversion factors, the estimation accuracy of soil organic matter content interval can be adjusted by the model optimized parameters, the estimation accuracy of soil organic matter content interval value of 12 checked samples is 91. 67%. It is showed that the interval hyper-spectral estimation of soil organic matter content is valid.
出处
《测绘科学技术学报》
CSCD
北大核心
2014年第6期593-597,602,共6页
Journal of Geomatics Science and Technology
基金
山东省自主创新专项项目(2012CX90202)
国家SRT项目(201410434099)
山东农业大学大数据中心项目(75006)
关键词
土壤有机质
高光谱
模糊识别
区间值估测
模型
soil organic matter
hyper-spectral
fuzzy recognition
interval estimation
model