Title :
Optimal supports for linear predictive models
Author :
Rajagopalan, Rajesh ; Orchard, Michael T. ; Ramchandran, Kannan
Author_Institution :
Dept. of Electr. Eng., Princeton Univ., NJ, USA
fDate :
12/1/1996 12:00:00 AM
Abstract :
The problem of finding the optimal set of causal pixels (support) for use in linear predictive models is addressed. After presenting counterexamples to popular intuitions about supports, a general result relating the distortion incurred with a small support to optimal coefficients of a larger support is derived. A geometrical interpretation is provided. Two algorithms that optimally increase/decrease support sizes by one at each step are presented. Experimental results illustrate the significant gains realized by the algorithms compared with commonly used supports
Keywords :
image processing; optimisation; prediction theory; causal pixels; distortion; geometrical interpretation; linear predictive models; optimal coefficients; optimal supports; support sizes; Autocorrelation; Autoregressive processes; Estimation error; Mean square error methods; Multidimensional systems; Nearest neighbor searches; Pixel; Predictive models; Signal processing; Yield estimation;
Journal_Title :
Signal Processing, IEEE Transactions on
