Optimal triplet halfband filter banks with k-regularity and image coding application

摘要: The triplet halfband filter bank structure is well known to be efficient for the design and implementation of a class biorthogonal wavelet banks. Previously, two extreme cases in characteristics including equiripple filters with no regularity maximally flat poor frequency selectivity have mainly been treated. This paper proposes an semi-definite programming (SDP) method linear phase finite impulse response (FIR) banks whose optimal prescribed order. By using matrix inequality (LMI) characterization trigonometric semi-infinite constraints, problem can exactly cast as SDP small number variables and, hence, solved efficiently. A example different orders provided validate proposed method. Finally, image coding performance presented.