A new and fast real-time implementation of 2-D DCT
A new and fast 4×4 DCT algorithm is proposed in this paper. This algorithm classifies the input 2-D pixel data into four groups, each of which is then further rearranged into 1-D DCT transform. Therefore, the computation of 2-D DCT can be implemented by four 1-D DCTs. As a result, the efficiency of 2-D DCT algorithm is dependent on the 1-D DCT algorithm adopted, and all the existing fast algorithms for 1-D DCT can be directly applied to further optimise the algorithm design. The proposed algorithm can also be extended to compute general 2-D DCT by a recursive procedure. Specific algorithm flowchart and design is also included and described.
- 1.Rao, K.R. and Yip, P. ‘Discrete cosine transform algorithms, advantages, and applications', Academic Press, INC. 1990, ISBN: 0-12-580203-x.Google Scholar
- 2.Hou, H.S. ‘A fast recursive algorithm for computing the discrete cosine transform', IEEE Trans. Acoust., Speech, Signal Processing, Vol. ASSP-35, No. 10, 1455–1461, 1987.Google Scholar
- 3.Lee, B.G. ‘A new algorithm to compute the discrete cosine transform’ IEEE Trans. Acoust., Speech, Signal Processing, Vol. ASSP-32, No.6, 1243–1245, Dec. 1984.Google Scholar
- 4.Feig, Ephraim, ‘Fast algorithms for the discrete cosine transform'. IEEE Trans. on Signal Processing, Vol. 40, No 9, 2174–2193, Sept. 1992.Google Scholar