Short and Medium-Term Prediction of Winter Wheat NDVI Based on the DTW–LSTM Combination Method and MODIS Time Series Data

Fa Zhao,Guijun Yang,Yang Meng,Xinlei Bu,Hao Yang,Yaohui Zhu,Shaoyu Han

doi:10.3390/rs13224660

Abstract

The normalized difference vegetation index (NDVI) is an important agricultural parameter that is closely correlated with crop growth. In this study, a novel method combining the dynamic time warping (DTW) model and the long short-term memory (LSTM) deep recurrent neural network model was developed to predict the short and medium-term winter wheat NDVI. LSTM is well-suited for modelling long-term dependencies, but this method may be susceptible to overfitting. In contrast, DTW possesses good predictive ability and is less susceptible to overfitting. Therefore, by utilizing the combination of these two models, the prediction error caused by overfitting is reduced, thus improving the final prediction accuracy. The combined method proposed here utilizes the historical MODIS time series data with an 8-day time resolution from 2015 to 2020. First, fast Fourier transform (FFT) is used to decompose the time series into two parts. The first part reflects the inter-annual and seasonal variation characteristics of winter wheat NDVI, and the DTW model is applied for prediction. The second part reflects the short-term change characteristics of winter wheat NDVI, and the LSTM model is applied for prediction. Next, the results from both models are combined to produce a final prediction. A case study in Hebei Province that predicts the NDVI of winter wheat at five prediction horizons in the future indicates that the DTW–LSTM model proposed here outperforms the LSTM model according to multiple evaluation indicators. The results of this study suggest that the DTW–LSTM model is highly promising for short and medium-term NDVI prediction.

Highlights

As an important development in the direction of agriculture in the future, precision agriculture can greatly improve productivity and input utilization efficiency
The aims of this study were as follows: (1) generate dynamic time warping (DTW) and long short-term memory (LSTM) deep neural network models capable of predicting the short- and mid-term normalized difference vegetation index (NDVI) of winter wheat with a high level of accuracy; (2) produce a method for combining the predictions of the DTW and LSTM models to increase the accuracy of predictions; and (3) assess the applicability, effectiveness, and advantages of the proposed DTW–LSTM combination model as a method for predicting the short- and mid-term NDVI with data from experiments conducted in Hebei province using five-year satellite time series data
The method used in this paper is a combined DTW and LSTM method based on the MODIS NDVI time series data and the winter wheat map described above

Summary

Introduction

As an important development in the direction of agriculture in the future, precision agriculture can greatly improve productivity and input utilization efficiency. The prediction of crop yield is an important focus of precision agriculture. Accurate yield prediction methods can provide support for good decision-making in agricultural planning, budgeting and input [1,2,3]. The normalized difference vegetation index (NDVI) is an indicator that reflects the greenness and productivity of vegetation, and it is closely related to the growth and yield of crops [4]. Compared with traditional remote sensing data, crop time series data have many advantages [9]. Crop time series data can reflect changes in long-term trends, seasonal cycles and random changes [10].

Objectives

Methods

Results

Conclusion