• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.1C
• /
• pp.35-44
• /
• 2004

In this paper, we design an efficient VLSI architecture of entropy coding algorithm in JPEG2000. Entropy coder is a context-based binary arithmetic encoder, and composed of a Context Extractor(CE) and an Arithmetic Coder(AC). We speed-up CE by skipping no-operation bits in coding passes, and AC is to be performed based on MQ coder. Because of using Qe value associated with each allowed context and probability estimation, MQ coder is a multiplication free coder that reduces computation loads and makes simple the structure of arithmetic coder. We have developed and synthesized the VHDL models of CE and AC pairs using Xilinx FPGA technology. The proposed architecture operates up to 30MHz.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.4B
• /
• pp.755-762
• /
• 2000

Entropy coders as a noiseless compression method are widely used as end-point compression for images so there have been many contributions to increase of entropy coder performance and to reduction of entropy coder complexity. In this paper, we propose some entropy coders based on binary forward classification (BFC). BFC requires overhead of classification but there is no change between the amount of input information and that of classified output information, which we prove this property in this paper. And using the proved property, we propose entropy coders which are Golomb-Rice coder after BFC (BFC+GR) and arithmetic coder with BFC(BFC+A). The proposed entropy decoders do not have further complexity Son BFC. Simulation results also show better performance than other entropy coders which have similar complexity to proposed coders.

• Journal of Advanced Navigation Technology
• /
• v.19 no.6
• /
• pp.614-622
• /
• 2015

Binary MQ arithmetic coding is widely used recently as a basic entropy coder in multimedia coding system. MQ coder esteems high in compression efficiency to be used in JBIG2 and JPEG2000. The importance of arithmetic coding is increasing after it is adopted as an unique entropy coder in HEVC standard. In the binary MQ coder, arithmetic approximation without multiplication is used in the process of recursive subdivision of range interval. Because of the MPS/LPS exchange activity happened in MQ coder, output byte tends to increase. This paper proposes an enhanced binary MQ arithmetic coder to make use of a lookup table for AQe using quantization skill in order to reduce the deficiency. Experimental results show that about 4% improvement of compression in case of JBIG2 for bi-level image compression standard. And also, about 1% improvement of compression ratio is obtained in case of lossless JPEG2000 coding. For the lossy JPEG2000 coding, about 1% improvement of PSNR at the same compression ratio. Additionally, computational complexity is not increasing.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1996.06a
• /
• pp.159-163
• /
• 1996

In this paper, a new image compression technique, which uses lifting scheme and pyramid vector quantization, is proposed. Lifting scheme is a new technique to generate wavelets and to perform wavelet transform, and pyramid vector quantization is a kind of vector quantization which dose not have codebook neither codebook generation algorithm. For the purpose of realizing more compression rate, an arithmetic entropy coder is used. Proposed algorithm is compared with other wavelet based image coder and with JPEG which uses DCT and adaptive Huffman entropy coder. Simulation results showed that the performance of proposed algorithm is much better than that of others in point of PSNR and bpp.

• 제어로봇시스템학회:학술대회논문집
• /
• 1989.10a
• /
• pp.984-986
• /
• 1989

101 coding scheme, one of sliding block coding techniques, provides practically attractive features in some compression applications for image sources such as facsimile. This paper presents a new simple model of 101 coder. The results show that the entropy of the output of the 101 coder can be reduced close to the rate distortion bound of a binary first-order markov source.

• Journal of Broadcast Engineering
• /
• v.13 no.5
• /
• pp.732-743
• /
• 2008

In H.264/AVC, 4$\times$4 block transform is used for intra and inter prediction instead of 8$\times$8 block transform. Using small block size coding, H.264/AVC obtains high temporal prediction efficiency, however, it has limitation in utilizing spatial redundancy. Motivated on these points, we propose a multi-dimensional transform which achieves both the accuracy of temporal prediction as well as effective use of spatial redundancy. From preliminary experiments, the proposed multi-dimensional transform achieves higher energy compaction than 2-D DCT used in H.264. We designed an integer-based transform and quantization coder for multi-dimensional coder. Moreover, several additional methods for multi-dimensional coder are proposed, which are cube forming, scan order, mode decision and updating parameters. The Context-based Adaptive Variable-Length Coding (CAVLC) used in H.264 was employed for the entropy coder. Simulation results show that the performance of the multi-dimensional codec appears similar to that of H.264 in lower bit rates although the rate-distortion curves of the multi-dimensional DCT measured by entropy and the number of non-zero coefficients show remarkably higher performance than those of H.264/AVC. This implies that more efficient entropy coder optimized to the statistics of multi-dimensional DCT coefficients and rate-distortion operation are needed to take full advantage of the multi-dimensional DCT. There remains many issues and future works about multi-dimensional coder to improve coding efficiency over H.264/AVC.

• Smart Media Journal
• /
• v.1 no.2
• /
• pp.6-11
• /
• 2012

In the HEVC standard, transform coefficient coding that affects the output bitstream directly is a core part of the encoder and it includes coefficient scanning and entropy coding. Recently, JCT-VC(Joint Collaborative Team on Video Coding) advances to HEVC Committee Draft (CD). In this paper, we explain HEVC transform coefficient coding and propose an efficient transform coefficient coding method considering statistics of transform coefficients in the intra frame coder. The proposed method reduces BD-Rate by up to 0.74%, compared to the conventional HEVC transform coefficient coding.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.08a
• /
• pp.189-192
• /
• 2000

A fuzzy controller for the evaluation of the quantization parameters in the H.263 coder has been introduced. We adopted a Mamdani fuzzy controller with centroid defuzzification. The inputs are variance, entropy, current motion vector and previous motion vector. This results is obtained a effective rate control technique using fuzzy Quantization.

• Journal of Korea Multimedia Society
• /
• v.11 no.11
• /
• pp.1480-1490
• /
• 2008

Design of a linear predictor and matching of an entropy coder is the art of lossless audio coding. In this paper, we use the covariance method and the Choleskey decomposition for calculating linear prediction coefficients instead of the autocorreation method and the Levinson-Durbin recursion. These results are compared to the polynomial predictor. Both of them, the predictor which has small prediction error is selected. For the entropy coding, we use the Golomb-Rice coder using the block-based parameter estimation method and the sequential adaptation method with LOCO-land RLGR. The proposed predictor and the block-based parameter estimation have $2.2879%{\sim}0.3413%$ improved compression ratios compared to FLAC lossless audio coder which use the autocorrelation method and the Levinson-Durbin recursion. The proposed predictor and the LOCO-I adaptation method could improved by $2.2879%{\sim}0.3413%$. But the proposed predictor and the RLGR adaptation method got better results with specific signals.

• Journal of the Institute of Convergence Signal Processing
• /
• v.11 no.2
• /
• pp.125-131
• /
• 2010

This paper presents the implementation of a EBCOT TIER-1 for JPEG2000 Encoder. JPEG2000 is new standard for the compression of still image for overcome the artifact of JPEG. JPEG2000 standard is based on DWT(Discrete Wavelet Transform) and EBCOT Entropy coding technology. EBCOT(Embedded block coding with optimized truncation) is the most important technology that is compressed the image data in the JPEG2000. However, EBCOT has the artifact because the operations are bit-level processing and occupy the harf of the computation time of JPEG2000 Compression. Therefore, in this paper, we present modified context extraction method for enhance EBCOT computational efficiency and implemented MQ- Coder as arithmetic coder. The proposed system is implemented by Verilog-HDL, under the condition of TSMC 0.25um ASIC library, gate counts are 30,511EA and satisfied the 50MHz operating condition.