Improved Yield Prediction of Winter Wheat Using a Novel Two-Dimensional Deep Regression Neural Network Trained via Remote Sensing.

Abstract

In recent years, the use of remotely sensed and on-ground observations of crop fields, in conjunction with machine learning techniques, has led to highly accurate crop yield estimations. In this work, we propose to further improve the yield prediction task by using Convolutional Neural Networks (CNNs) given their unique ability to exploit the spatial information of small regions of the field. We present a novel CNN architecture called Hyper3DNetReg that takes in a multi-channel input raster and, unlike previous approaches, outputs a two-dimensional raster, where each output pixel represents the predicted yield value of the corresponding input pixel. Our proposed method then generates a yield prediction map by aggregating the overlapping yield prediction patches obtained throughout the field. Our data consist of a set of eight rasterized remotely-sensed features: nitrogen rate applied, precipitation, slope, elevation, topographic position index (TPI), aspect, and two radar backscatter coefficients acquired from the Sentinel-1 satellites. We use data collected during the early stage of the winter wheat growing season (March) to predict yield values during the harvest season (August). We present leave-one-out cross-validation experiments for rain-fed winter wheat over four fields and show that our proposed methodology produces better predictions than five compared methods, including Bayesian multiple linear regression, standard multiple linear regression, random forest, an ensemble of feedforward networks using AdaBoost, a stacked autoencoder, and two other CNN architectures.

Plant phenotyping relevance

圃場内の冬コムギ収量という植物形質をリモートセンシングから推定するCNN手法を開発し、複数手法との交差検証で性能比較しているため、表現型推定法が中心である。

Code and data availability

The supplied blocks describe winter wheat yield datasets (Sentinel-1 imagery, nitrogen rate, topographic rasters, yield monitor maps) and the Hyper3DNetReg model, but contain no public data deposit, code repository, or availability statement. Field names were hidden for privacy, and no author URL for data or code is in

No evidence-backed public reproduction asset is currently recorded.