Detection of Small Target Based on Morphological Filters

To efficiently decrease the number of false alarms in detection of infrared small targets with morphological filters, a post processing algorithm is proposed in this paper.Morphological filters such as hit-or-miss and modified top-hat transform has been used to enhance and detect infrared dim small target embedded in cluttered background. However, falsealarms will be increased in real scenario with heavy clutter. To overcome this problem, fast post-processing algorithm based onRobinson guard filter is presented here. Contrast feature is used in this algorithm to improve the thresholding technique of theprevious work using morphological filters. The experimental results on several infrared image sequences of small targets withheavy clutter background show that our algorithm is efficient.

Keywords:

detection , multiple targets , hit-or-miss transformetion , top-hat transformation , robinson guard filter