Abstract
This paper introduces a new color image superresolution algorithm in an adaptive, robust M-estimation framework. Using a robust error norm in the objective function, and adapting the estimation process to each of the low-resolution frames, the proposed method effectively suppresses the outliers due to violations of the assumed observation model, and results in color superresolution estimates with crisp details and no color artifacts, without the use of regularization. Experiments on both synthetic and real sequences demonstrate the superior performance over using the L2 and L1 error norms in the objective function.
Highlights
Image super-resolution (SR) is a popular research area for producing high-resolution (HR) images with better details
Most of the methods that have been developed recently in the literature for color image SR in the context of Mestimation [28,29,30,31] use the L2 or the L1 error norm in the data fidelity term of the objective function, and incorporate an additional color regularization term to help in dealing with the outliers and ill-posedness of the SR problem
To improve the robustness of SR reconstruction, we propose the use of robust error norms in the data fidelity term of the objective function, in particular, the robust error norms which correspond to the class of M-estimators known as redescending M-estimators [1,2,3,4]
Summary
Image super-resolution (SR) is a popular research area for producing high-resolution (HR) images with better details. Most image SR algorithms assume a mathematical model for the imaging process, which could have generated the sequence of LR frames from the unknown HR image. These models are only approximations to reality, and model violations often occur because of the approximate nature of the model itself, because of inaccuracies in its parameter estimation (such as blur and motion parameters) and because of accidental scene changes. El-Yamany et al [10] developed an adaptive M-estimation scheme using the robust Lorentzian error norm in the data fidelity term, without regularization. The proposed approach was first introduced in [13]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
