Comparison of inversion method of maize leaf area index based on UAV hyperspectral remote sensing

The current inversion method of maize leaf area index has the problems of long time-consuming inversion, high energy consumption, and low fitting coefficient between the prediction result and the actual result. For these problems, an inversion method of maize leaf area index based on UAV hyperspectral remote sensing is proposed in this paper. Acquisition of spectral image of maize leaf using ASD Field SpecPro FR Field Hyperspectral instrument and CCD Array detector in UAV remote Sensing system. Denoising and segmentation of corn leaf images using stationary wavelets, improved Snake and PSO methods.. The improved Snake model is used to achieve coarse convergence of target image contour after denoising. Through particle swarm optimization iterative algorithm, the optimal image segmentation point is found and the image segmentation is achieved. Based on the results of image denoising segmentation, the expressions of modified chlorophyll absorption ratio, normalized difference spectral index, and ratio-type spectral index are obtained. The correlation between the three indices and the maize leaf area index was analyzed. Finally, the maize leaf area index was obtained by using ratio-type spectral index. Experiments show that the proposed method has a comprehensive performance, and has a strong advantage over the current method.