Using Combined Vegetation Indices to Monitor Leaf Chlorophyll Content in Winter Wheat Based on Hj-1a/1b Images

摘要

This article investigates the relationships between leaf chlorophyll content (LCC) and these combination variables derived from vegetation indices, which are extracted from HJ-1A/1B images. The combined models ? new measures of monitoring LCC, are compared to single vegetation index model. The results demonstrate that normalization combination for normalized difference vegetation index (NDVI) and green normalized difference vegetation index (GNDVI), namely N (NDVI, GNDVI), is feasible to monitor winter wheat LCC at jointing stage (node formation). R 2 and RMSE are 0.861 and 0.345, respectively, which are more ideal than those of single vegetation index model. The accuracy increases by 3.4%. Ratio combination for NDVI and GNDVI, namely R (NDVI, GNDVI), is feasible to monitor LCC at booting stage. R2 and RMSE are 0.616 and 0.208, respectively, which are more ideal than those of single vegetation index model. The accuracy increases by 15.1%. Difference combination for NDVI and GNDVI, namely D (NDVI, GNDVI), is feasible to monitor LCC at anthesis. R2 and RMSE are 0.694 and 0.409, respectively, which are more ideal than those of single vegetation index model. The accuracy increases by 13%. In conclusion, the combined models can provide a new method for accurately monitoring crop growth conditions in the future.