基于改进小波阈值的条纹图像去噪研究

Research on Fringe Image Denoising Based on Improved Wavelet Threshold

摘要: 为消除条纹结构光图像中的噪声干扰，提出了基于小波变换的改进阈值函数，并结合条纹图像的特征信息分布，选择合适的估计阈值，对条纹图像进行去噪。实验结果表明，小波阈值去除条纹图像的白高斯噪声过程中，相比选用传统的阈值函数，本文提出的改进阈值函数使峰值信噪比和结构相似度更高。

Abstract: In order to eliminate the noise interference in the fringe structured light image, an improved threshold function based on wavelet transform was proposed, and combined with the feature information distribution of the fringe image, an appropriate estimation threshold was selected to denoise the fringe image. Experimental results show that in the process of removing white Gaussian noise from fringe images with wavelet threshold, the improved threshold function proposed in this paper makes the peak signal-to-noise ratio and structural similarity higher than the traditional threshold function.