#### DMCA

## Two-Dimensional Discrete Cosine Transform (2002)

### Citations

235 |
A Fast Computational Algorithm for the Discrete Cosine Tranfsorm
- Wen-Hsiung, Smith, et al.
- 1977
(Show Context)
Citation Context ...umption at the algorithmic, architectural, and circuit levels. Similar to the second design, chip C used row-column decomposition to extend the 1-D DCT to 2-D. The DA version of the Chen Fast 1-D DCT =-=[7]-=- was chosen in chip C. This is the same algorithmicscombination used in the DUT. Similar to chips A and B, chip C [10] also used custom circuitry tosachieve low power. Adders were all designed with a ... |

38 |
JPEG: Still Image Data Compression.
- Pennebaker, Mitchell
- 1992
(Show Context)
Citation Context ... be captured from just the summation of the lower order values. ThesDCT coefficients needed for the reconstruction process can be found mathematically from the following JPEG appropriate DCT formulas =-=[2]-=-: Forward DCT in 1-dimension on 8 samples:sThe two dimensional DCT is merely a product of terms between a horizontal 1-D DCT and a vertical 1-D DCT. S u( ) C u( ) 2 ----------- s x( ) uπ 16 ----- 2x 1... |

28 |
Quantifying and enhancing power awareness of VLSI systems
- Bhardwaj, Min, et al.
- 2001
(Show Context)
Citation Context ... with one another, the amount of computation scaled down.s4.1 Power Awareness Distributed arithmetic implementations lend themselves well to power awareness schemes. Powersawareness, as formulated in =-=[13]-=-, pertains to the ability of a system to dynamically scale downspower consumption based on operating conditions.While conventional designs operate under 33 worst-case operating scenarios, power-aware ... |

20 |
Computation of an odd-length DCT from a real-valued DFT of the same length
- Heideman
- 1992
(Show Context)
Citation Context ...-based algorithm of [4] shows increased performance in throughput, latency, and turnaroundstime. Optimization with respect to these parameters is not the focus of the proposed project. AsDFT approach =-=[5]-=- calculates the odd-length DCT, which is not applicable to this project since thesdesign must be compatible with JPEG standards. S v u,( ) C u( ) 2 ----------- C v( ) 2 ----------- s y x,( ) uπ 16 ---... |

10 |
High Throughput CORDIC-Based Systolic Array Design for the Discrete Cosine Transform
- Hsiao, Chen, et al.
- 1995
(Show Context)
Citation Context ...ry of 2-D DCT implementation is indirect computation through other transforms--most commonly, the Discrete Hartley Transform (DHT) and the Discrete Fourier Transform (DFT). ThesDHT-based algorithm of =-=[4]-=- shows increased performance in throughput, latency, and turnaroundstime. Optimization with respect to these parameters is not the focus of the proposed project. AsDFT approach [5] calculates the odd-... |

7 |
Low Power Data-Dependent Transform Video and Still Image Coding
- Xanthopoulos
- 1999
(Show Context)
Citation Context ...o the LUT. The key elements required to implement the DA are a 16-element LUT and decoder, an adder/ subtractor, and a shifter. These elements are grouped together in a ROM Accumulate (RAC)sstructure =-=[10]-=- as shown below (figure 3.4).sA shift register inputs one column of input bits persclock cycle to the LUT. It begins by inputting the most significant column of bits and rotates tosfinally input the l... |

2 |
A cost effective architecture for 8x8 2-D DCT/IDCT using direct method
- Lee, Chen, et al.
- 1997
(Show Context)
Citation Context ...lications and 896 additions. Fast algorithms exploit the symmetry within the DCT tosachieve dramatic computational savings.sThere are three basic categories of approach for computation of the 2-D DCT =-=[3]-=-. The first category of 2-D DCT implementation is indirect computation through other transforms--most commonly, the Discrete Hartley Transform (DHT) and the Discrete Fourier Transform (DFT). ThesDHT-b... |

2 |
A Fast Algorithm for 2-D
- Cho, Yun, et al.
- 1991
(Show Context)
Citation Context ...d using the results of aspolynomial transform. Computational complexity is greatly reduced, but regularity is sacrificed.sInstead of the 16 1-D DCTs used in the conventional row-column decomposition, =-=[6]-=- uses all realsarithmetic including 8 1-D DCTs, and stages of pre-adds and post-adds (a total of 234 additions)sto compute the 2-D DCT. Thus, the number of multiplications for most implementations sho... |

1 | et al., “A 0.9V, 150MHz, 10-mV, 4mm2, 2-D Discrete Cosine Transform Core Processor with Variable Threshold-Voltage (VT) Scheme - Kuroda - 1996 |