Small Target Detection In Infrared Clutter Using Dark Channel Prior And Improved Local Entropy

In this article, we propose an infrared background clutter suppression technique using a similar method with the haze removal algorithm based on dark channel prior. The proposed method introduce some modifications to make the original haze removal algorithm can work in gray-scale infrared image in real-time. In addition, after the background suppression, we address a efficient algorithm for small target detection, which incorporate Box-Filtering technique into traditional small infrared target detecting based on LEA. This improved method reduces a large number of redundant calculations in traditional Local Entropy algorithm by applying the iterative operations instead of using traditional approach. We provide detailed analysis for time complexity and the results of simulation experiments with real world infrared images in the final of this work.