自适应小波阈值与平均算法去噪在拉曼光谱中的应用

拉曼光谱分析中,噪声的存在常影响分析的准确度和检测限.现有滤波方法在光谱信号除噪方面有种种缺陷.基于Dohono提出的小波阈值去噪,使用一种自适应小波阈值函数滤噪法并与平均算法相结合,实现信号与噪音的分离.该方法除噪完全,即使对信噪比小于1的高噪声信号也能够很好地保留信号的细节,获取满意的处理结果.

The Application of Adaptive Wavelet Threshold Denoising and Average Arithmetic in Raman Spectra

The existence of random noise is inevitable in the Raman spectral analysis,and it makes a negative influence on the accuracy of analysis and detection limit.There are many disadvantages to the present methods of wavelet filtering in spectral signals.A novel adaptive wavelet threshold function denoising method with average arithmetic based on Dohono's wavelet shrinkage method was developed.By using this method separation is achieved between signal and noise and denoise completely,and even for signals with a highsignal-to-noise ratio less than one,this method can also be employed to keep the main edges of signal and get rid of noise.A good effect in spectral processing can be achieved.