Recent baseline extraction and correction techniques are based on Penalized Least Square method; which is focused on two mains parameters: weight vector and smooth parameters estimation. Weight vector is computed iteratively based on the difference between original signal and the ith extracted baseline, when the smooth parameters are generally computed empirically. The drawback of these techniques is that the algorithm associated for baseline optimization; which mainly overestimated if the signal is below a fitted baseline and under estimated when the signal is above a fitted baseline. In this paper, we proposed an efficient algorithm for robust baseline extraction; in which the optimal weight vector is computed based on logic distribution function; and, the smooth parameters using PCA method. The new algorithm has been extended to existing extraction methods. Simulations results have shown the effectiveness of the propose algorithm, and the advantage of using multi-smooth parameters for automatic baseline removal.

Keywords: Baseline extraction method, Spectral analysis, logic distribution function, Penalized Least Square method, PCA, Optimization