复杂高光谱图像噪声去除数学模型的设计研究

Design of mathematical model for noise removal of complex hyperspectral images

传统图像去噪模型去噪效果不理想,为了获得更优质的复杂高光谱图像,改善复杂高光谱图像质量,设计了复杂高光谱图像去噪的数学模型。首先设计复杂高光谱图像的退化和复原模型,并对复杂高光谱图像进行分解与重构,然后利用积分法消除复杂高光谱图像中的随机噪声,最后从主观和客观两个方面对复杂高光谱图像去噪效果进行评价测试。结果表明,相对于其他复杂高光谱图像去噪模型,本模型的复杂高光谱图像视觉效果更佳,大幅度提升了复杂高光谱图像信噪,获得了更优的复杂高光谱图像去噪效果。

The denoising effect of traditional image denoising model is not ideal.In order to obtain higher quality complex hyperspectral image and improve the quality of complex hyperspectral image,a mathematical model of complex hyperspectral image denoising is designed.Designs the degradation and restoration model of hyperspectral image,introduces modern mathematical theory to decompose and reconstruct hyperspectral image,and uses integral method to eliminate random noise in hyperspectral image.The hyperspectral image denoising effect is evaluated and tested from both subjective and objective directions.Compared with other hyperspectral image denoising models,the hyperspectral image visual effect of this model is better,the hyperspectral image signal-to-noise is greatly improved,and a better hyperspectral image denoising effect is obtained.