Abstract:

In order to achieve effective detection of weak targets in infrared complex background, a filtering algorithm combining morphological Top-hat transform and improved nonlinear diffusion (based on Perona-Malik’s (PM) research) model was proposed to enhance infrared Weak target signal, suppress complex background and noise. The method firstly uses the Top-hat operator in morphological filtering to enhance the target of the image, and then uses the modified PM filter to further filter the morphologically filtered image to achieve the purpose of suppressing the background to highlight the target, and finally the dim target detection is realized by threshold segmentation. The comparison experiments show that the proposed algorithm can effectively suppress the background and edge of the infrared weak target image at low signal-to-noise ratio, increase the signal-to-noise ratio of the image by 20%, and increase the detection ability by 40% in the original algorithm.