Infrared small target detection via adaptive M-estimator ring top-hat transformation. (April 2021)

• Record Type: Journal Article
• Authors: Deng, Lizhen; Zhang, Jieke; Xu, Guoxia; Zhu, Hu
• Is Part Of: Pattern recognition. Volume 112(2021)
• Publication Date: 2021-04
• Subjects: Infrared small target detection -- Top-hat transformation -- M-estimator -- Local entropy; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2020.107729
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: An adaptive ring top-hat transformation for infrared target detection is presented. Designing a structural element based on M-estimator and ring shape. A novel local entropy is proposed further to incorporate with top-hat transformation. The proposed algorithm yields high detection probability under complex background. Abstract: Top-Hat transformation is an essential technology in the field of infrared small target detection. Many modified Top-Hat transformation methods have been proposed based on the different structure of structural elements. However, these methods are still hard to handle the dim targets and complex background. It can be summarized as two reasons, one is that the structural elements cannot suppress the background adaptively due to the fixed value of structural elements in image. Another is that simple structural element cannot utilize the local feature for target enhancement. To overcome these two limitations, a special ring Top-Hat transformation based on M-estimator and local entropy is proposed in this paper. First, an adaptive ring structural element based on M-estimator is used to suppress the complex background. Second, a novel local entropy is proposed to weight structural element for capturing local feature and target enhancement. Finally, a comparison experiment based on massive infrared image data (more than 500 infrared target images) is done. And the results demonstrate that the proposed algorithm acquires better performance compared with some recent methods.