Prediction of Potential Yied from Spectral Reflectance of Winter Wheat Based on Nonlinear Dynamics Reconstruction

摘要：

For reasons of less accurate Potential yield prediction in winter wheat only with NDVI method, we propose a way to combine NDVI and local dynamics reconstruction for deterministic nonlinear yield prediction. The ensemble time series consisting of 17 individual crop samples in validation dataset are used to validate this proposed method, some nonlinear analysis such as correlation dimension are conducted to analyze underlying dynamics in the short yield time series which is converted from reflectance data of wheat to virtual yield data with NDVI prediction model. Then we reconstruct the local dynamics of the vaidation dataset with the time delay embedding technique.Finally the Jacobian prediction method based on local dynamics reconstruction is introduced to realize the prediction of winter wheat potential yield in the phase space. Prediction results show the applicability of this method, and also indicate that Jacobian prediction can get more accurate for crop potential yield prediction than NDVI prediction obviously.

关键词： Nonlinear dynamics reconstruction Chaotic analysis NDVI Potential yield prediction

作者: ZHANG Junning LI Minzan

作者单位: Key Laboratory of Modern Precision Agriculture System Integration Research,Ministry of Education, China Agricultural University, Beijing 100083, China

会议类型: 国际会议

会议名称: 2008 International Conference on IAEA(2008农业信息化、自动化与电气化国际会议)

会议地点: 镇江

会议语种:英文

页码: 573-578

在线出版日期: 2008-11-01（万方平台首次上网日期，不代表论文的发表时间）