Nonparametric Complex Background Prediction Algorithm Using FCM Clustering for Dim Point Infrared Targets Detection

A nonparametric background prediction algorithm using fuzzy c-means (FCM) clustering is proposed to enhance the detection of dim small infrared targets in image data. The target of interest is assumed to have a very small spatial spread, and is obscured by heavy background clutter. The input image data is firstly segmented using FCM clustering, and then the nonparametric regressive method is applied to predict background in each cluster respectively. Subsequently the background is subtracted from the input data, leaving components of the target signal in the residual noise. Experiment results show better detecting performance for the output data by the algorithm of this paper than by other traditional methods