Monitoring Lodging Extents of Maize Crop Using Multitemporal GF-1 Images

Maize crop lodging is a recurrent phenomenon which results in significant reduction of grain yield and quality addition to the impediment mechanical harvesting. The large-scale monitoring maize important for production policy adjustment agricultural insurance compensation. In this article, we derived variety features from multitemporal GaoFen-1 (GF-1) images before after lodging. We screened most sensitive spectrum, texture, vegetation index monitor recursive feature elimination method based on cross-validation mutual information were compared obtain optimal combination extents crop. random forest classifier was used classify extents. showed that indices included difference reflectance blue, green, red bands, normalized index, ratio enhanced difference, mean value blue band, green band. total accuracy classification 87.50%, Kappa coefficient 0.83 testing samples. Based multiple GF-1 lodging, can be monitored large scale.