High performance scalable image compression with EBCOT

Taubman, David

doi:10.1109/icip.1999.817132

Cited by 253 publications

(310 citation statements)

References 14 publications

Supporting

Mentioning

310

Contrasting

Order By: Relevance

“…Table 2 illustrates the PSNR results of three algorithms over a wide range of compression ratios. The three algorithms are our proposed algorithm, the EBCOT algorithm of JPEG2000 [21], and Martin's algorithm [6]. From Table 2, we can see that compression with our algorithm always yields a higher PSNR than compression with Martin's algorithm, and a lower (about 0.4-0.5 dB) PSNR than that of the JPEG2000 algorithm.…”

Section: Resultsmentioning

confidence: 79%

“…Then we compare these results with those of JPEG2000 and Martin's algorithm. The kernel algorithm of JPEG2000 is EBCOT [21]. JPEG2000 is currently the most popular algorithm for image coding.…”

Section: Resultsmentioning

confidence: 99%

“…The transforms in which JPEG2000 are used are the 9/7 and 5/3 scalar wavelet transforms. Compared with the famous SPIHT algorithm [21], we know this: the PSNR of the image that is reconstructed with the JPEG2000 algorithm is lower than that of the image that is reconstructed by the SPIHT algorithm with Figure 10. Set of conditioning contexts used by the arithmetic encoder: (a) adjacent pixels, (b) parent pixels.…”

Section: Resultsmentioning

confidence: 99%

See 2 more Smart Citations

Context- and HVS-Based Multiwavelet Image Coding Using SPIHT Framework

Yü

Zhou

2005

Circuits Syst Signal Process

View full text Add to dashboard Cite

In this paper, multiwavelets are considered in the context of image compression based on the human vision system (HVS). First, selecting the BSA (4/4)* filters, a twodimensional image is transformed with our proposed algorithm I. Second, we apply HVS coefficients into the subbands of the transformed image. Third, we split the coefficients into two parts: the significance map and residue map. Then a new modified set partitioning in hierarchical trees (SPIHT) algorithm is proposed to encode the significance map. Fourth, algorithm III is presented for coding the residue map. Finally, we adopt context-based adaptive arithmetic coding to encode the bit stream. We also provide some experimental results proving that multiwavelets are worth studying and compare them with those of other multiwavelet and JPEG2000 algorithms.

show abstract

Section: Resultsmentioning

confidence: 79%

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Context- and HVS-Based Multiwavelet Image Coding Using SPIHT Framework

Yü

Zhou

2005

Circuits Syst Signal Process

View full text Add to dashboard Cite

show abstract

“…The model given by (13) and (14) is an LP problem and thus is easily solvable by the modified simplex method [9, p. 175]. In this model, the decision variables x i,b are assumed to be continuous, and hence the optimal solutions x * i,b can take any fractional or real values within the specified bounds [0, u i,b −u i,b−1 ].…”

Section: Methodsmentioning

confidence: 99%

Real-Time Optimized Error Protection Assignment for Scalable Image and Video over Wireless Channels

Muhit

Chuah

2008

Circuits Syst Signal Process

View full text Add to dashboard Cite

A new error protection assignment scheme with applications to real-time wireless multimedia transmission is presented. The proposed scheme exploits the structure of scalable sources to ensure optimal assignment. This novel approach recasts the nonlinear optimization problem into a linear one, and then further remodels it into a discrete programming problem, thereby reducing the computational complexity dramatically. Furthermore, the proposed algorithm does not impose any requirement on the convexity of the source; i.e., it can equally be applied on a convex or nonconvex source. Results show that the described method facilitates a significant complexity reduction with respect to existing schemes, while exhibiting almost equivalent performance.

show abstract

“…There are two kinds of watermarks: visible and invisible. For this paper, we are only interested in the invisible watermarks [1,[8][9][10][11][12]. An invisible watermark is an image embedded in a host image that cannot be easily detected by the raw human visual.…”

Section: Introductionmentioning

confidence: 99%

A simple and efficient watermark technique based on JPEG2000 Codec

Chen

2004

Multimedia Systems

View full text Add to dashboard Cite

In this paper we propose an effective watermark scheme for embedding and extracting based on the JPEG2000 Codec process. Our embedding algorithm applies the torus automorphisms (TA) technique to break up and scramble a watermark. The scrambled watermark was embedded into the quantized bitstreams of JPEG2000 before the entropy coding stage. Distortion reduction (DR) was applied to the compressed image to lessen image degradation caused by the embedding process. Our watermark scheme is simple and easy to implement. Furthermore, it is robust to attacks like blurring, edge enhancement, and other image processing manipulations.

show abstract

High performance scalable image compression with EBCOT

Cited by 253 publications

References 14 publications

Context- and HVS-Based Multiwavelet Image Coding Using SPIHT Framework

Context- and HVS-Based Multiwavelet Image Coding Using SPIHT Framework

Real-Time Optimized Error Protection Assignment for Scalable Image and Video over Wireless Channels

A simple and efficient watermark technique based on JPEG2000 Codec

Contact Info

Product

Resources

About