Automatic determination of the spread parameter in Gaussian smoothing

The Gaussian filter is a very good filter for smoothing signals or images. The amount of smoothing depends on the value of the spread parameter (i.e., the standard deviation) of the Gaussian function. A heuristic and efficient algorithm to automatically determine the spread parameter for smoothing a given one-dimensional signal is proposed in this paper. The proposed algorithm uses an iterative process to compute a single spread parameter for smoothing the given signal. Only a small number of iterations are needed for smoothing most signals using the algorithm. Moreover, we extend the single-parameter Gaussian smoothing algorithm to multi-parameter smoothing for characterizing local variations of the signal. The algorithms are computationally fast and need no threshold. Experimental results show that the algorithms are indeed efficient and effective.