A novel algorithm for detecting dim point target in infrared images based on MRF

摘要

In this paper a dim point target detection algorithm based on Markov Random Field (MRF) is proposed. Firstly, the Min-Difference Filter (MDF) is applied to suppress the background in each single frame. Then an adaptive segmentation process based on 3D MRF is used to detect targets in image sequence. Unlike the traditional 3D MRF, the spatial MRF is modified and a Point-to-Region mapping according to the specific situation is proposed. Experiment results show that the proposed algorithm allows to detect IR point targets in complex backgrounds.