Method for feature extraction and small target detection based on infrared polarization

Tiao-wen QIU,Yan ZHANG,Wei-ping YANG,Ji-cheng LI
DOI: https://doi.org/10.3969/j.issn.1001-5078.2014.10.018
2014-01-01
Abstract:To solve the problem of clutter interference in infrared target detection,a novel method of feature extraction and small target detection based on the infrared polarization information is proposed.Firstly,according to the differ-ences of polarization characteristics of artificial target and natural background,a model of infrared polarization informa-tion of targets and the background is established.The components of intensity and polarization information are analyzed as well as the extraction and enhancement of the target polarization information.Then,the three theories of the stokes vector,partially polarized light decomposition and variable polarization are combined to extract target polarization fea-ture,suppress background clutter and eliminate random noise,thus achieve detection of infrared small tar-get.Experimental results on simulated data and comparisons with three existed methods demonstrate the effectiveness and reliability of the proposed method,which can be used against infrared camouflage concealment or infrared decoy technology,and has great application value.
What problem does this paper attempt to address?