基于小波变换的共焦拉曼图像去噪方法

Confocal Raman image denoising method based on wavelet transform

针对现有共焦拉曼光谱显微图像噪声去除不足的问题,提出一种基于小波变换的拉曼图像全面去噪方法。首先对拉曼光谱图像进行小波变换处理,然后再对拉曼图像进行双边滤波处理,进而实现拉曼光谱显微图像中噪声干扰的抑制。理论分析和实验表明:与现有拉曼光谱显微图像去噪方法相比,所提方法使拉曼光谱显微图像的峰值信噪比和结构相似度分别改善了92.99%和197%,其有效提高了拉曼光谱图像的信噪比和物质成分分析的准确率。

In order to solve the problem that the existing methods reduce the noise of the confocal Raman spectroscopy microscopy images insufficiently, a method based on wavelet transform is proposed to remove the noise of the confocal Raman spectral images comprehensively. The method processes the confocal Raman spectral images by wavelet transform firstly, and then performs bilateral filtering on the Raman image to suppress noise interference in the confocal Raman spectrum images. Both the theoretical analysis and experiments show that compared with the exiting methods, this method improves respectively the peak signal-to-noise ratio and structural similarity of the confocal Raman spectral image by 92.99% and 197%. It improves the signal-to-noise ratio of the confocal Raman spectral image and the accuracy of material composition analysis effectively.