Title :
Constrained regularization methods for superresolution
Author :
Connolly, T.J. ; Lane, R.G.
Author_Institution :
Dept. of Electr. & Electron. Eng., Canterbury Univ., Christchurch, New Zealand
Abstract :
We apply constrained Tikhonov-Miller regularization to superresolution. The effective solution of this problem necessitates the imposition of a positivity constraint on the conjugate gradient method. This results in a fast projection based quadratic programming method for image superresolution. Difficulties with existing algorithms for this problem are also examined
Keywords :
conjugate gradient methods; image resolution; quadratic programming; conjugate gradient method; constrained Tikhonov-Miller regularization; constrained regularization methods; fast projection based quadratic programming method; image superresolution; positivity constraint; Algorithm design and analysis; Clustering algorithms; Constraint theory; Convergence; Equations; Fourier transforms; Gradient methods; Image restoration; Quadratic programming;
Conference_Titel :
Image Processing, 1998. ICIP 98. Proceedings. 1998 International Conference on
Conference_Location :
Chicago, IL
Print_ISBN :
0-8186-8821-1
DOI :
10.1109/ICIP.1998.999056
