A Detection Method of Infrared Target Based on the Ridgeline Extraction

Target detection in infrared imagery is a challenging problem due to extremely low resolution, low signal-to-noise ratio (SNR) and poor target visibility of real images. There has always been a high false alarm probability for the traditional detection method. In this paper, a novel detection method for removing the false alarm is proposed, which consists of the following steps. First, multi-scale characteristic views models are obtained from a 3D object reference model, which will be used for preprocessing and target detection. A detection method is presented for extraction of the ridge contour line used to get a region of interest. Then, the potential target regions are obtained by preprocessing the real image using morphological grayscale operators and iterative segmentation. Lastly, the target is detected by matching the feature models information from the potential target regions within the region of interest. Experimental results are shown to demonstrate the robustness and efficiency of the proposed method from forward looking infrared image sequences with complicated background.