基于光谱分析的莲雾叶片全氮含量模型的研究

Study on Model for Total Nitrogen of Wax Apple Leaves Based on the Spectrum by Analysis

摘要: 本研究以5年和6年莲雾为研究对象，以田间试验为依托，利用便携式光谱仪测定莲雾叶片的光谱反射率，采用逐步回归法分析不同氮素水平下莲雾叶片原始光谱反射率与叶片氮素含量的关系，确定敏感波段、构建基于原始光谱反射的模型，最后，经过对模型的优选与检验，得出最佳的估测模型。研究结果如下：莲雾叶片全氮含量的原始光谱的敏感波长为510、560、1200 nm，相关系数分别为−0.888**、0.873**、0.997**，对莲雾叶片全氮与原始光谱的敏感波段采用逐步回归分析法建立定量监测模型并进行检验，最终确定的最佳预测模型为Y = 2.629 X1200 + 4.104，研究结果将为莲雾营养诊断、长势监测和估产等领域提供理论依据。

Abstract: With Syzygium samarangense as the research object and based on the field experiment, hyper-spectral reflectance was measured by CropScan MSR16 in leaves of 5 and 6 years’ Syzygium samarangense. This study has adopted stepwise regression to analyze the relationship between the original spectral reflectance and nitrogen concentration of in the leaf blade under different nitrogen levels to determine the sensitive wave bands and to build the model based on the original spectral reflection. Finally, the best estimate model has been attained after optimizing and testing. The results showed as follows: the sensitive wave lengths of the original spectrum under three levels of the content of nitrogen are 510, 560 and 1200 nm, and the correlation coefficients are 0.888**, 0.873**, 0.997** respectively. The quantitative monitoring model between different level of the content of nitrogen and the original spectrum band has been established with stepwise regression analysis and checking. Ultimately, the best prediction model is Y = 0.63 X1200 + 4.104. The research results will provide the theoretical basis for nutrition diagnosis, condition monitoring and yield estimation of Syzygium samarangense.