论文部分内容阅读
土壤有机质与水分均对土壤反射光谱特征有显著的影响,但其作用机理难以定量描述。通过对黑龙江省典型黑土区土壤野外高光谱反射率的测定,研究了该区土壤的光谱反射特征;利用BP神经网络方法,以土壤有机质与水分数据作为输入层,以土壤光谱反射率一阶微分作为输出层,建立黑土有机质与水分的高光谱预测模型,并对模型的稳定性和预测能力进行检验。结果表明:土壤有机质与土壤表层0~20 cm含水量之间具有显著的相关性,相关系数为0.59;1570 nm波段处的一阶微分为输出层的模型精度最高,RMES达到0.017,平均绝对误差为0.014,平均相对误差为0.110;模型检验结果表明建立的BP神经网络模型具有良好的稳定性;土壤有机质、含水量对土壤光谱反射率的综合作用得到较准确的描述,可以用于野外土壤有机质与水分的速测。
Both soil organic matter and water have a significant effect on soil reflectance spectral characteristics, but its mechanism of action is difficult to describe quantitatively. The spectral reflectance characteristics of soils in typical black soil area of Heilongjiang Province were studied. The spectral reflectance characteristics of soils were studied. By using BP neural network method, the soil organic matter and moisture data were taken as input layer, As the output layer, the hyperspectral prediction model of organic matter and water in black soil was established, and the stability and predictive ability of the model were tested. The results showed that there was a significant correlation between soil organic matter and water content of 0-20 cm on the soil surface, with a correlation coefficient of 0.59. The model with the first derivative at the 1570 nm band as the output layer had the highest accuracy with RMES 0.017 and mean absolute error And the average relative error is 0.110. The model test results show that the established BP neural network model has good stability. The combined effect of soil organic matter and water content on soil spectral reflectance is more accurately described and can be used for the field soil organic matter Quick test with moisture.