Monitoring of growth of wheat’s height using time series analysis of synthetic aperture radar (SAR) images and the corresponding parameters

Abstract

In today’s word, precision agriculture is one of the practical solutions in increasing agricultural products and timely monitoring to increase food security in today's world. Today, remote sensing technology and GIS are used for this purpose. Classification of various crops, especially wheat, in order to monitor its growth stages in multispectral images due to limitations such as dependence on weather conditions and lack of night imaging. At the same time, SAR images are capable of capturing images in all weather conditions, as well as day-to-day imaging, overcoming the limitations of optical images. Due to the identification of the phenomenon based on the geometry, hardness and orientation of objects, it provides us with a lot of information. Therefore, the main purpose of this study is to evaluate the feasibility of using RADAR images to monitor the growth cycle of wheat. For this study, a bipolar radar image of Sentinel 1 with polarization (VV and VH) will be used to obtain plant growth parameters such as height and biomass. Then, the parameters obtained in order to analyze the time series to understand the plant growth cycle will be performed unsupervised in the Google Earth Engine system. The results will show that band C in the early stages of wheat growth is most sensitive to wheat height and have appropriate information. Field data will then be used to assess accuracy. Final result demonstrates that VV polarization is far better indicator of Wheat growth in early stage, however, VH polarization is more sensitive to middle stage of plant growth cycle. Moreover, neither VV nor VH can detect Wheat growth cycle in their last stage. Both VV and VH showed high connection with the peak of Wheat growth in early May. In fact, it is said that it would be better way to use L band and longer wavelength for Wheat growth in its last stages.