Minimizing the entropy of quantized post-interpolation residuals for hierarchical image compression

Adaptive parameterized interpolation for hierarchical compression of images is investigated. I propose an approach for optimizing the parameters of an adaptive interpolator, based on minimizing the estimation of the compressed data size (entropy of quantized post-interpolation residues). Based on the proposed approach, I develop an adaptive parameterized interpolator and propose a recursive procedure for estimating its parameters. I perform an experimental comparison of the proposed interpolator with averaging interpolators and an adaptive interpolator based on minimizing the sum of the interpolation errors. The advantage of the proposed interpolator in the size of compressed data is demonstrated at various maximum errors.