Search

The fast rate-optimal rate allocation algorithm used selects a subchannel and channel code rate based on signal-to-noise ratios of the subchannels and gives unequal error protection.

A joint source-channel decoding scheme is proposed that is based on JPEG2000 codestreams and LDPC channel codes. The results have shown that this method needs very less iterations and has improved overall system performance.

An algorithm is presented that efficiently exploits inter- and intrasensor correlation for the purpose of power-constrained distributed transmission of sensor-network imagery. Results demonstrate that gains in network lifetime up to 114e obtained when using the suggested algorithm with lossless compression, and much larger gains are achieved when lossy compression is employed.

Two leaky- bucket rate allocation methods that provide constant quality video under buffer constraint are proposed. The goal is to allocate the capacity of the channel between sequences to achieve constant quality across all sequences and the results demonstrate substantial benefits from the use of Motion JPEG2000.

A fully scalable image coder is presented and the lossless and lossy performances are investigated.

Research describing an improved image restoration algorithm is presented. Using original and diffraction-limited image pairs test results show that the nonlinear algorithm produces super-resolution.

Building from uniform quantization thresholds, the authors present a new method of trellis coded quantization named universal trellis coded quantization (UTCQ). The UTCQ's symmetry to trellis supersets, and its use to improve the picture performance of particular images at low birates are discussed.

Techniques are presented for allowing progressive transmission for situations involving the application of trellis coded quantization to wavelet coefficients, along with a process for approximately inverting trellis coded quantization if least significant bits are absent.

A data compression method is presented based on specific bit allocations within synthetic aperture radar systems. The process includes applying inverse Fourier transform and aperture weighting operations to radar system phase history data, then attaching a gain factor to each sample, which directs bit allocation.

Image compression research has established a near-lossless imaging system which centers around differential pulse code modulation techniques (DPCM). It reduces quantized prediction error sequences to a minimum and sets up contexts for a conditional prediction error model. The DPCM trellis is illustrated, together with encoding rate tables.

An investigation of various classification schemes applied to the subband coding of images found that the dependence between classification maps was used in encoded by constraining the classification maps. The complexities of the iterative algorithms used in the schemes made analysis of their functionalities very difficult. Simulation, however, demonstrated the value of the schemes and their algorithms. The results of the simulation indicate that the fourth scheme generally performed better than the other schemes.

A hyperspectral image coding approach that involves the use of the two-dimensional discrete cosine transform (2-D DCT) and entropy-constrained predictive trellis coded quantization (ECPTCQ) is developed. The 2-D DCT is employed for the transformation of all non-overlapping eight-by-eight blocks of each band. ECPTCQ is then utilized for encoding the transform coefficients in the spectral dimension. Iterative optimization algorithms are applied to codebook design and rate allocation.

A predictive image coder uses adaptive recursive interpolative DPCM along with entropy-constrained trellis-coded quantization and optimal rate allocation to reduce decoder complexity. It is able to provide the same range of signal-to-noise ratios as advanced transform coders making it a high-performance adaptive RIDPCM-ECTCQ image coder.

The use of entropy-constrained trellis-coded quantization for encoding monochrome and color images was studied. The technique, which achieves efficient data compression, was implemented on images decomposed into frequency subbands by the discrete wavelet transform. A performance evaluation, using quadrature mirror filters and nine-tap spline filters, was carried out. It reproduced images without fuzziness or high-frequency background noise.

A methodology for visually lossless compression of monochrome stereoscopic 3D images is proposed. Visibility thresholds are measured for quantization distortion in JPEG2000. These thresholds are found to be functions of not only spatial frequency, but also of wavelet coefficient variance, as well as the gray level in both the left and right images. To avoid a daunting number of measurements during subjective experiments, a model for visibility thresholds is developed. The left image and right image of a stereo pair are then compressed jointly using the visibility thresholds obtained from the proposed model to ensure that quantization errors in each image are imperceptible to both eyes. This methodology is then demonstrated via a particular 3D stereoscopic display system with an associated viewing condition. The resulting images are visually lossless when displayed individually as 2D images, and also when displayed in stereoscopic 3D mode. [ABSTRACT FROM AUTHOR]

A technique for identifying image blur using vector quantizer encoder distortion is presented. This technique needs a set of training images blurred by each of the candidate blur functions to design a corresponding vector quantizer codebook.

Due to exponential growth in image sizes, visually lossless coding is increasingly being considered as an alternative to numerically lossless coding, which has limited compression ratios. This paper presents a method of encoding color images in a visually lossless manner using JPEG2000. In order to hide coding artifacts caused by quantization, visibility thresholds (VTs) are measured and used for quantization of subband signals in JPEG2000. The VTs are experimentally determined from statistically modeled quantization distortion, which is based on the distribution of wavelet coefficients and the dead-zone quantizer of JPEG2000. The resulting VTs are adjusted for locally changing backgrounds through a visual masking model, and then used to determine the minimum number of coding passes to be included in the final codestream for visually lossless quality under the desired viewing conditions. Codestreams produced by this scheme are fully JPEG2000 Part-I compliant. [ABSTRACT FROM AUTHOR]

Scanning orders of bitplane image coding engines are commonly envisaged from theoretical or experimental insights and assessed in practice in terms of coding performance. This paper evaluates classic scanning strategies of modern bitplane image codecs using several theoretical-practical mechanisms conceived from rate–distortion theory. The use of these mechanisms allows distinguishing those features of the bitplane coder that are essential from those that are not. This discernment can aid the design of new bitplane coding engines with some special purposes and/or requirements. To emphasize this point, a low-complexity scanning strategy is proposed. Experimental evidence illustrates, assesses, and validates the proposed mechanisms and scanning orders. [ABSTRACT FROM AUTHOR]

This work addresses the transmission of pre-encoded JPEG2000 video within a video-on-demand scenario. The primary requirement for the rate allocation algorithm deployed in the server is to match the real-time processing demands of the application. Scalability in terms of complexity must be provided to supply a valid solution by a given instant of time. The FAst rate allocation through STeepest descent (FAST) method introduced in this work selects an initial (and possibly poor) solution, and iteratively improves it until time is exhausted or the algorithm finishes execution. Experimental results suggest that FAST commonly achieves solutions close to the global optimum while employing very few computational resources. [ABSTRACT FROM PUBLISHER]

Bitplane coding is a common strategy used in current image coding systems to perform Iossy, or Iossy-to-lossless, compression. There exist several studies and applications employing bitplane coding that require estimators to approximate the distortion produced when data are successively coded and transmitted. Such estimators usually assume that coefficients are uniformly distributed in the quantization interval. Even though this assumption simplifies estimation, it does not exactly correspond with the nature of the signal. This work introduces new estimators to approximate the distortion produced by the successive coding of transform coefficients in bitplane image coders, which have been determined through a precise approximation of the coefficients' distribution within the quantization intervals. Experimental results obtained three applications suggest that the proposed estimators are able to approximate distortion with very high accuracy, providing a significant improvement over state-of-the-art results. [ABSTRACT FROM AUTHOR]

A fast rate-optimal rate allocation algorithm is proposed for parallel transmission of scalable images in multi-channel systems. Scalable images are transmitted via fixed-length packets. The proposed algorithm selects a subchannel, as well as a channel code rate for each packet, based on the signal-to-noise ratios (SNR5) of the subchannels. The resulting scheme provides unequal error protection of source bits and significant gains are obtained over equal error protection schemes. An application of the proposed algorithm to JPEG2000 transmission shows the ad- vantages of exploiting differences in SNRs between subchannels. Multiplexing of multiple sources is also considered, and additional gains are achieved by exploiting information diversity among the sources. [ABSTRACT FROM AUTHOR]

The high compression efficiency and various features provided by JPEG2000 make it attractive for image transmission purposes. A novel joint source/channel coding scheme tailored for JPEG2000 is proposed in this paper to minimize the end-to-end image distortion within a given total transmission rate through memoryless channels. It provides unequal error protection by combining the forward error correction capability from channel codes and the error detection/localization functionality from JPEG2000 in an effective way. The proposed scheme generates quality scalable and error-resilient codestreams. It gives competitive performance with other existing schemes for JPEG2000 in the matched channel condition case and provides more graceful quality degradation for mismatched cases. Furthermore, both fixed-length source packets and fixed-length channel packets can be efficiently formed with the same algorithm. [ABSTRACT FROM AUTHOR]

Provides information on the presentation of a novel technique for image restoratin based on nonlinear interpolative vector quantization (NLVQ). Information on the image restoration; Detailed information on nonlinear VQ image restoration; Conclusion reached.