Author_Institution :
Dept. of Electr. Eng., Univ. of Maranhao, San Luis, Brazil
Abstract :
The discrete Fourier transform (DFT) of a real sequence f[x, y] of size N×N, where N=2n, can be computed by a two-dimensional (2-D) FFT of size N/4, or smaller if f[x, y] is known to have certain symmetries. This paper presents theorems that identify the symmetry in f[x, y] based on the depth of the quadtree to expedite 2-D FFT computation of coherent digital images. In principle, it establishes that if the quadtree of f[x, y] has maximum depth k<n where K=2n , then the DFT can be computed by a 2-D FFT of size K/2. An algorithm is given, and its performance analyzed. Finally, applications are considered in transform coding systems and lossy compression of images
Keywords :
data compression; discrete Fourier transforms; fast Fourier transforms; image coding; image sequences; quadtrees; transform coding; 2D FFT computation; algorithm; applications; digital images; discrete Fourier transform; lossy image compression; performance; quadtree; real sequence; symmetry; transform coding systems; Algorithm design and analysis; Data structures; Digital images; Discrete Fourier transforms; Helium; Image coding; Performance analysis; Sparse matrices; Transform coding; Two dimensional displays;