Research on a Quantitative Assessment Model Based on Visual Perception in Low-Altitude Remote Sensing

In order to evaluate the recovery effect of the visible light band in the low-altitude remote sensing, the sensitivity and significance of the visual perception of the restored image are selected and optimized by the perceptual characteristics of the human visual system. Finally, the effective detail retention ability, Structure information, multi - scale similarity and visual fidelity. A kind of image - based image fog quantification evaluation model QAMVP is proposed based on visual perception. The experimental results show that the model has a strong judgment force on the object group with obvious difference in the image quality evaluation. The accuracy rate of the sample group is 92%. In the aspect of visual perception,, The MSE of the model index curve is obviously lower than the comprehensive evaluation method and the PSNR measurement method, which indicates that the model has certain advantages in the degree of visual perception and the correlation with the visual perception of the human eye. The results of the study can provide theoretical basis and practical guidance for the objective evaluation of the visible image in the low-altitude remote sensing.