The soil organic matter content grey relationship inversion pattern based on hyper-spectral technique

摘要

Purpose - The purpose of this paper is to establish the grey-weighted relationship prediction pattern of the soil organic matter content spectral inversion under the uncertainties between soil organic matter contents and spectral characteristics and the theory of grey system. Design/methodology/approach - At first, according to grey-weighted distance, a new grey relationship model is presented. Second, in order to make full use of the information of grey relationship sequences, the maximum grey relationship discrimination principle is improved and then the soil organic matter content spectral inversion pattern is put forward based on weighted grey recognition theory. A numeric example of Hengshan County in Shanxi Province is also computed in the last part of the paper. Findings - The results are convincing: not only that soil organic matter content spectral inversion pattern based on the weighted grey recognition theory is valid, but also the model's prediction accuracy is higher; the sample's average prediction accuracy is 94.917 per cent. Practical implications - The method exposed in the paper can be used at soil organic matter content hyper-spectral inversion and even for other similar forecast problems. Originality/value - The paper succeeds in realising both prediction pattern and application of soil organic matter content hyper-spectral inversion by using the newest developed theories: weighted grey recognition theory.