An Infrared Small Target Detection Algorithm Utilizing Joint Local Contrast Measure

The effectiveness of infrared small target detection technology depends on high detection accuracy. However, when dealing with complex backgrounds or targets close to the edges of jamming objects, existing algorithms typically encounter significant challenges, such as high false alarm rates and low detection probabilities. To address this issue, we propose a novel measurement scheme utilizing joint local contrast measure based on background enhancement and edge suppression (JLCM-BEES), in which a new scanning window is employed to compute the joint local contrast of infrared images, with the focus on enhancing the target while suppressing the background, a new filtering kernel strengthens the background and suppresses the target, and two related structural elements are used for dilation and differencing operations on the infrared image, retaining the target while effectively suppressing edges with strong contrast. A new fusion method is introduced to combine the three obtained saliency maps, followed by the separation of the target from the background using an adaptive thresholding algorithm. Experimental results indicate that the proposed algorithm demonstrates superior detection capability and robustness when facing complex conditions and low signal-to-clutter ratios, providing new insights for the development of infrared small target detection technology.