Monitoring Barley Growth Condition with Multi-scale Remote Sensing Images

Crop growth condition monitoring at regional scale with remote sensing data has been widely implemented. The normal method is extracting biophysical and biochemical parameters, and then setting thresholds for these parameters to grade different levels of crop growth. In which, as the parameters inversion has scale effects based on different remote sensing observations with different spatial scales, it is difficult to setting the threshold at multi-spatial scales. To achieve space consistency for multi-scale crop growth monitoring results, we constructed two new vegetation indexes for crop growth monitoring, and then proposed a new crop growth grading system. We constructed two new crop growth indicators, i.e., Crop Growth Monitoring Index 1 (CGMI1), and Crop Growth Monitoring Index 2 (CGMI2), based on Leaf Area Index (LAI) and Canopy Chlorophyll Density (CCD). Compared with the existed crop growth indicators, these two new growth indicators could provide a much more comprehensive description of the characteristics of crop growth status from the aspects of crop structure and biochemical conditions. To achieve the space consistency of crop growth monitoring, we constructed a new crop growth grading system based on multiple spatial resolution satellite images. Firstly, we proposed a spatial adaptive threshold selection method by integrating with data histogram and Gaussian distribution theory for thresholds selection based on the statistical analysis of CGMI1 and CGMI2, then to strengthen robustness of threshold selecting on multi-scale. Moreover, we carried out research on crop growth monitoring and ranking based on the selected thresholds of CGMI1 and CGMI2 from the aspects of crop canopy morphology structure (large, medium, and small) and crop canopy biological activity (strong, middle, and weak). Taking barley as our research object, three multi-source and multi-scale remote sensing images are obtained during the jointing-booting stage of barley, which include Advanced Land Observing Satellite-Advanced Visible and Near Infrared Radiometer type 2 (ALOS-AVNIR2) image, Small Remote Sensing Satellite Constellations A Star-CCD2 (HJ 1A-CCD2) image, and the 8-day composite MODIS Surface Reflectance Product (MOD09A1). Experimental numerical results showed better space consistency for crop growth monitoring based on multiple spatial scale dataset (ALOS, HJ, and MODIS). The new proposed crop growth indicators CGMI1 and CGMI2 based on LAI and CCD to both consider the crop morphology structure and biological activity. And the new growth grading rules provide a spatial adaptive threshold selection algorithm to keep the space consistency when mapping different crop growth grading. Theoretical analysis and numerical experiments fully confirmed the new system, not only effectively enhance the crop growth evaluation, but also revealing better results on the space consistency with multi-scale data.