Blind super-resolution using sparse line gradient prior for passive millimeter wave image

Aiming at the low resolution problem of Passive Millimeter Wave Image, a blind super-resolution algorithm based on sparse line gradient (SLG) prior is proposed. Firstly, the line gradient is defined to modeling the specialized prior, which is in line with the characteristics of PMMW image than custom gradient. Then the super-resolution problem under MAP framework is optimized in an alternative iteration progress. Experimental results demonstrate that for the millimeter wave image blind super-resolution, the proposed algorithm is superior to other traditional method.