论文部分内容阅读
软阈值消噪是信号消噪中的标准算法.理论上,软阈值方法在最小最大误差方面是近似最优的.研究表明,通过结合系数消噪和软阈值方法,可以达到更低的误差下界.由于离散小波是非平移不变的,因而重构过程中会出现人工噪声.为了避免这个问题,采用了具有平移不变性的二进小波变换.实验结果表明,文中所提算法的消噪结果具有更高的信噪比和更光滑的外观.
Soft threshold denoising is the standard algorithm in signal denoising.In theory, the soft threshold method is approximately optimal in the minimum and maximum error.The research shows that by combining the coefficient denoising and soft threshold method, lower error bounds can be achieved Since the discrete wavelet transform is non-parallel, artificial noise will appear in the reconstruction process.To avoid this problem, a binary wavelet transform with translational invariance is used.The experimental results show that the denoising result of the proposed algorithm has Higher signal to noise ratio and smoother appearance.