An infrared small target detection framework based on local contrast method. (September 2016)

• Record Type: Journal Article
• Authors: Cui, Zheng; Yang, Jingli; Li, Junbao; Jiang, Shouda
• Is Part Of: Measurement. Volume 91(2016)
• Page Start: 405
• Page End: 413
• Publication Date: 2016-09
• Subjects: Small target detection -- Modulation map -- Human visual system -- High speed local contrast method -- Machine learning; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2016.05.071
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: A novel framework inspired by human visual system is presented. Modulation map is defined. Abnormal regions are detected in the modulation map. Improve local contrast method to high speed local contrast method. Abstract: Small-target detection in infrared imagery used in forward looking infrared is an important task in remote sensing fields. It is important to improve the detection capabilities such as detection rate, false alarm rate and speed. In this letter, a novel approach inspired by human visual system (HVS) is presented. First, block compressed sampling theory was used to compress the image and obtain the modulation map. Then, small abnormal regions in the modulation map were detected by high speed local contrast method and defined as candidate targets. Experimental results show the proposed algorithm pursue good performance in detection rate, false alarm rate and speed simultaneously.