论文部分内容阅读
叶面积指数(leaf area index,LAI)是重要的生物物理参数,亦是各种生态模型、生产力模型以及碳循环研究等的重要生物物理参量,因此具有重要的研究意义。通过分析大量实测数据,选用归一化植被指数(normalized difference vegetation index,NDVI)和比值植被指数(ratio vegetation index,RVI)、主成分分析(principcal component analysis,PCA)、神经网络(neural network NN)三种方法对大豆使LAI进行了估算,比较分析了三种方法的估算结果。研究结果表明,植被指数法(NDVI,RVI),主成分分析,神经网络方法LAI都取得了较为理想的结果,验证模型的确定性系数分别达0.758和0.753,0.954,0.899,其中主成分分析方法和神经网络方法精度较高,主成分分析方法验证模型的稳定性更好,其验证模型的RMSE为0.267,明显低于两个植被指数(NDVI和RVI的RMSE分别为0.594和0.616)和神经网络(RMSE=0.413)。当叶面积指数较小时,植被指数能够较好地去除土壤、大气等条件影响,并精确估算LAI;当叶面积指数较大时,主成分分析能够弥补植被指数饱和的缺陷,得到很好的LAI估算效果。神经网络受LAI大小的影响效果居中,其对叶面积指数较小和较大时具有一致的估算效果,具有较好的应用潜力。
Leaf area index (LAI) is an important biophysical parameter. It is also an important biophysical parameter of various ecological models, productivity models and carbon cycle research. Therefore, LAI is of great research significance. By analyzing a large number of measured data, normalized difference vegetation index (NDVI) and ratio vegetation index (RVI), principalcal component analysis (PCA), neural network NN Three methods for soybean LAI were estimated, comparative analysis of the three methods of estimation. The results show that the LAI of the vegetation index method (NDVI, principal component analysis, neural network method) has achieved good results, the validation coefficient of the validation model were 0.758 and 0.753,0.954,0.899, respectively, of which the principal component analysis And the neural network method are more accurate. The principal components analysis method is more stable. The RMSE of the validation model is 0.267, which is significantly lower than that of two vegetation indices (RMSE of NDVI and RVI are 0.594 and 0.616, respectively) and neural network (RMSE = 0.413). When leaf area index is small, the vegetation index can remove the influence of soil and atmosphere conditions and estimate LAI accurately. When the leaf area index is larger, PCA can make up for the deficit of vegetation index saturation and get a good LAI Estimate the effect. The effect of LAI size is the most important effect of neural network. It has the same estimation effect when the leaf area index is small and large, which has a good potential for application.