Scalar Quantization Research Articles

Zero-Delay Joint Source-Channel Coding Using Hybrid Digital-Analog Schemes in the Wyner-Ziv Setting

This paper studies zero-delay joint source-channel coding (JSCC) of Gaussian sources over additive white Gaussian noise (AWGN) channels in the Wyner-Ziv scenario, that is, when there is side information about the source available only to the receiver. The proposed encoding scheme is based on hybrid digital-analog (HDA) transmission in a zero-delay fashion. Specifically, to achieve zero-delay, after applying scalar quantization to the source, the properly scaled analog information (quantization error) is superimposed on the scaled digital information (quantized source), and then transmitted. At the receiver side, several reconstruction schemes are analyzed. It is shown that all the schemes, when optimized, are superior to existing uncoded, digital, or hybrid transmission methods. It is also shown that the proposed HDA transmission may result in increased robustness to channel and/or side information mismatch compared to uncoded transmission. Observing that the bandwidth-matched Wyner-Ziv problem is intimately related to the coding scenario without side information but with bandwidth expansion factor 2, the proposed scheme is applied to the latter problem. It is shown that this application results in a better performance than the well-known inverse spiral mapping.

A GA-based quantisation design scheme for ECG data compression with linear distortion control

Wavelet-based ECG data compression with multi-level adjustment mechanism is disadvantageous for reconstruction error control due to multi-toone mapping between multi-level quantisation scales and distortion. Linear distortion compression performance with easy error control is crucial for quality guarantee desired in diagnosis use. In this paper, a systematic approach is proposed for wavelet-based quantisation scheme design. This approach combines a Genetic Algorithm (GA) and Three-Dimensional (3-D) curvefitting technology to form a cascaded design process. The former is used to obtain high Compression Ratio (CR) and the latter aims to design a linear distortion control. Two kinds of ECG signals saved in MIT arrhythmia and ST change databases were used for the evaluation of compression performance and distortion control. The experimental results show that the design method can obtain linear distortion characteristic with high compression performance being kept.

Improvement of wireless transmission system performance for EEG signals based on development of scalar quantization

Abstract Advancement of wireless technology leads to some developments in current wireless electroencephalography. Through improving the transmission method of brainwaves, it would be possible to bring more convenience for the patients in need and give this opportunity to others for discovering other aspects of the amazing brainwave. What has been proposed in this study is a new type of adjustable backward quantization method which exploits the nature of the brainwave signal. This method is based on the nature of the captured brainwave and its quantization boundary changes based on the amplitude of each EEG captured signal. The proposed quantization scheme has been analyzed with uniform and Gaussian distribution of quantization level. Consequently, the Backward Gaussian Quantization with Adjustable Boundary and two Word Memories beside the Backward Uniform Quantization with Adjustable Boundary and two Word Memories are introduced by this experiment. In addition, the performance of wireless transmission system and the proposed quantizer’s efficiency for very low frequency (up to 100 Hz) and amplitude EEG signal have been noticed. With doing so, we simulated the transmitter and receiver by MATLAB® software. To model the medium, channel was assumed as Additive White Gaussian Noise (AWGN). Meanwhile analysis is done for the whole wireless system performance in terms of transmission range, compared with current available wireless transmission systems on the market. It should be noticed that the transmission range of the proposed wireless transmission system is compared to the transmission range of current wireless EEG systems when there is no obstacle between transmitter and receiver. Furthermore, some relevant parameters to evaluate the quality of the proposed quantization method were examined. To sum up, the proposed quantization schemes show considerable performance in terms of Quantization Rate for constant MSQE and SQNR in comparison with Uniform Quantization method and the achieved transmission range of our wireless system by using this method is higher than available wireless EEG systems on market.

Stabilizing Nonuniformly Quantized Compressed Sensing With Scalar Companders

This paper addresses the problem of stably recovering sparse or compressible signals from compressed sensing measurements that have undergone optimal nonuniform scalar quantization, i.e., minimizing the common <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$\ell _{2}$</tex></formula> -norm distortion. Generally, this quantized compressed sensing (QCS) problem is solved by minimizing the <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$\ell _{1}$</tex></formula> -norm constrained by the <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex Notation="TeX">$\ell _{2}$</tex></formula> -norm distortion. In such cases, remeasurement and quantization of the reconstructed signal do not necessarily match the initial observations, showing that the whole QCS model is not consistent. Our approach considers instead that quantization distortion more closely resembles heteroscedastic uniform noise, with variance depending on the observed quantization bin. Generalizing our previous work on uniform quantization, we show that for nonuniform quantizers described by the “compander” formalism, quantization distortion may be better characterized as having bounded weighted <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$\ell _{p}$</tex></formula> -norm ( <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$p \geqslant 2$</tex></formula> ), for a particular weighting. We develop a new reconstruction approach, termed Generalized Basis Pursuit DeNoise (GBPDN), which minimizes the <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$\ell _{1}$</tex></formula> -norm of the signal to reconstruct constrained by this weighted <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$\ell _{p}$</tex></formula> -norm fidelity. We prove that, for standard Gaussian sensing matrices and <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$K$</tex> </formula> sparse or compressible signals in <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$ \BBR ^{N}$</tex></formula> with at least <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$\Omega ((K \log N/K)^{p/2})$</tex></formula> measurements, i.e., under strongly oversampled QCS scenario, GBPDN is <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$\ell _{2}-\ell _{1}$</tex></formula> instance optimal and stable recovers all such sparse or compressible signals. The reconstruction error decreases as <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex Notation="TeX">$O(2^{-B}/\sqrt {p+1})$</tex></formula> given a budget of <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex Notation="TeX">$B$</tex></formula> bits per measurement. This yields a reduction by a factor <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex Notation="TeX">$\sqrt {p+1}$</tex></formula> of the reconstruction error compared to the one produced by <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$\ell _{2}$</tex></formula> -norm constrained decoders. We also propose an primal-dual proximal splitting scheme to solve the GBPDN program which is efficient for large-scale problems. Interestingly, extensive simulations testing the GBPDN effectiveness confirm the trend predicted by the theory, that the reconstruction error can indeed be reduced by increasing <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$p$</tex> </formula> , but this is achieved at a much less stringent oversampling regime than the one expected by the theoretical bounds. Besides the QCS scenario, we also show that GBPDN applies straightforwardly to the related case of CS measurements corrupted by heteroscedastic generalized Gaussian noise with provable reconstruction error reduction.

$\ell_{2}$ Optimized Predictive Image Coding With $\ell_{\infty}$ Bound

In many scientific, medical, and defense applications of image/video compression, an [Symbol: see text]∞ error bound is required. However, pure[Symbol: see text]∞-optimized image coding, colloquially known as near-lossless image coding, is prone to structured errors such as contours and speckles if the bit rate is not sufficiently high; moreover, most of the previous [Symbol: see text]∞-based image coding methods suffer from poor rate control. In contrast, the [Symbol: see text]2 error metric aims for average fidelity and hence preserves the subtlety of smooth waveforms better than the ∞ error metric and it offers fine granularity in rate control, but pure [Symbol: see text]2-based image coding methods (e.g., JPEG 2000) cannot bound individual errors as the [Symbol: see text]∞-based methods can. This paper presents a new compression approach to retain the benefits and circumvent the pitfalls of the two error metrics. A common approach of near-lossless image coding is to embed into a DPCM prediction loop a uniform scalar quantizer of residual errors. The said uniform scalar quantizer is replaced, in the proposed new approach, by a set of context-based [Symbol: see text]2-optimized quantizers. The optimization criterion is to minimize a weighted sum of the [Symbol: see text]2 distortion and the entropy while maintaining a strict [Symbol: see text]∞ error bound. The resulting method obtains good rate-distortion performance in both [Symbol: see text]2 and [Symbol: see text]∞ metrics and also increases the rate granularity. Compared with JPEG 2000, the new method not only guarantees lower [Symbol: see text]∞ error for all bit rates, but also it achieves higher PSNR for relatively high bit rates.

Design of the adaptive piecewise uniform scalar quantizer with lossless coder and Golomb-Rice code on the output, for signals with Gaussian distribution

In this paper a new compression model for signals with Gaussian distribution is presented. It consists of a piecewise uniform scalar quantizer, a lossless coder and Golomb-Rice code. Forward adaptation of this proposed model is done and constant signal-to-quantization noise ratio (SQNR) in wide range of input variances is obtained. Joined design of lossy and lossless coder is done, achieving optimal performances. Our model has small complexity since all components of the model (piecewise uniform quantizer, lossless coder and Golomb-Rice code) are very simple for realization. As an example, this model is applied on the speech signal, and it is shown that the G.712 standard is satisfied with bit-rate decrease of 1.09 bits/sample, compared to the classic uniform quantizer.

Detection of frame deletion for digital video forensics

The abundance of digital video forms a potential piece of evidence in courtrooms. Augmenting subjective assessment of digital video evidence by an automated objective assessment helps increase the accuracy of deciding whether or not to admit the digital video as legal evidence. This paper examines the authenticity of digital video evidence and in particular it proposes a machine learning approach to detecting frame deletion. A number of discriminative features are extracted from the video bit stream and its reconstructed images. The features are based on prediction residuals, percentage of intra-coded macroblocks, quantization scales and reconstruction quality. The importance of these features is verified by using stepwise regression. Consequently, the dimensionality of the feature vectors is reduced using spectral regression where it is shown that the projected features of unaltered and forged videos are nearly separable. Machine learning techniques are used to report the true positive and false negative rates of the proposed solution. It is shown that the proposed solution works for detecting forged videos regardless of the number of deleted frames, as long as it is not a multiple of the length of a group of pictures. It is also shown that the proposed solution is applicable for the two modes of video compression, variable and constant bitrate coding.

Scalar Quantization for Estimation: From An Asymptotic Design to a Practical Solution

In this paper, we study an asymptotic approximation of the Fisher information for the estimation of a scalar parameter using quantized measurements. We show that, as the number of quantization intervals tends to infinity, the loss of Fisher information induced by quantization decreases exponentially as a function of the number of quantization bits. A characterization of the optimal quantizer through its interval density and an analytical expression for the Fisher information are obtained. A comparison between optimal uniform and nonuniform quantization for the location and scale estimation problems shows that nonuniform quantization is only slightly better than uniform quantization. As the optimal quantization intervals are shown to depend on the unknown parameters, by applying adaptive algorithms that jointly estimate the parameter and set the thresholds in the location and scale estimation problems, we show that the asymptotic results can be approximately reached in practice using only 4 or 5 quantization bits.

Meta-heuristic algorithms for optimized network flow wavelet-based image coding

Optimal multipath selection to maximize the received multiple description coding (MDCs) in a lossy network model is proposed. Multiple description scalar quantization (MDSQ) has been applied to the wavelet coefficients of a color image to generate the MDCs which are combating transmission loss over lossy networks. In the networks, each received description raises the reconstruction quality of an MDC-coded signal (image, audio or video). In terms of maximizing the received descriptions, a greater number of optimal routings between source and destination must be obtained. The rainbow network flow (RNF) collaborated with effective meta-heuristic algorithms is a good approach to resolve it. Two meta-heuristic algorithms which are genetic algorithm (GA) and particle swarm optimization (PSO) have been utilized to solve the multi-objective optimization routing problem for finding optimal routings each of which is assigned as a distinct color by RNF to maximize the coded descriptions in a network model. By employing a local search based priority encoding method, each individual in GA and particle in PSO is represented as a potential solution. The proposed algorithms are compared with the multipath Dijkstra algorithm (MDA) for both finding optimal paths and providing reliable multimedia communication. The simulations run over various random network topologies and the results show that the PSO algorithm finds optimal routings effectively and maximizes the received MDCs with assistance of RNF, leading to reduce packet loss and increase throughput.

A Bit Allocation Method for Sparse Source Coding

In this paper, we develop an efficient bit allocation strategy for subband-based image coding systems. More specifically, our objective is to design a new optimization algorithm based on a rate-distortion optimality criterion. To this end, we consider the uniform scalar quantization of a class of mixed distributed sources following a Bernoulli-generalized Gaussian distribution. This model appears to be particularly well-adapted for image data, which have a sparse representation in a wavelet basis. In this paper, we propose new approximations of the entropy and the distortion functions using piecewise affine and exponential forms, respectively. Because of these approximations, bit allocation is reformulated as a convex optimization problem. Solving the resulting problem allows us to derive the optimal quantization step for each subband. Experimental results show the benefits that can be drawn from the proposed bit allocation method in a typical transform-based coding application.

Performance analysis of compressed sensing with uniform scalar quantisation

In compressed sensing, the sampling rate can be greatly reduced based on the sparsity of the original signal; however, quantisation is one key aspect in signal acquirement. This article studies the scalar quantisation and its impacts on the compressed sensing. We consider the case in which both the original and the measurement signals are quantised or discrete, which widely exists in the practical applications such as the finite precision signals in the computer and the finite-set original signal in communication systems. We derive the recovery probability bound and the necessary measurement number bound for compressed sensing with scalar quantisation and simulate three modified compressed sensing recovery algorithms. Simulations show that coarse quantisation can be applied in compressed sensing for discrete or quantised sparse signal with sampling and storage efficiency.

Rate control for unbalanced multiple description video streaming

Unbalanced multiple description coding (U-MDC) of video signals aims to obtain different coding rates for transmission over channels with asymmetric available bandwidth. This paper proposes a new rate control method, based on Multiple Description Scalar Quantisation (MDSQ), to generate unbalanced descriptions for asymmetric multipath video streaming. A new set of index assignment tables combined with an extension of the ρ model is devised to generate two unbalanced descriptions with controlled rate, using U-MDC under the general coding framework of H.264/AVC. This work demonstrates that a linear relationship between the rate and percentage of zeros in the transform coefficients of each description is maintained across different description domains, i.e., when moving from the single description domain (SDC) into the MDC domain. The simulation results show that rate–distortion performance of U-MDC is better than its equivalent balanced MDC. The proposed rate control algorithm exhibits high accuracy in achieving different target bitrates and unbalanced ratios between two descriptions. Improved performance is also achieved in video streaming over asymmetric transmission paths, where each description rate can be adapted according to different channel conditions.

A Low-Complexity 3.6kbps Speech Coding Algorithm Based on Mixed Excitation

A low-complexity 3.6kb/s speech coding algorithm based on mixed excitation is presented in this paper. It uses the parameter encoding and mixed excitation technology to ensure the quality of speech. Through adopting the scalar quantization of Line Spectrum Frequency (LSF), the algorithm reduces the storage and computational complexity. Meanwhile, improved frame type with dynamic Unvoiced/Voiced (U/V) thresholds make a reduction of the traditional U/V decision error and the sudden transformation of U/V frame. A modified bit allocation table is introduced and the PESQ-MOS test shows that the synthetic speech quality has been improved and reached the quality of communication, especially for high frequency female speakers with new frame type.

Information Rates of Densely Sampled Data: Distributed Vector Quantization and Scalar Quantization With Transforms for Gaussian Sources

This paper establishes rates attainable by several lossy schemes for coding a continuous parameter source to a specified mean-squared-error distortion based on sampling at asymptotically large rates. First, a densely sampled, spatiotemporal, stationary Gaussian source is distributively encoded. The Berger-Tung bound to the distributed rate-distortion function and three convergence theorems are used to obtain an upper bound, expressed in terms of the source spectral density, to the smallest attainable rate at asymptotically large sampling rates. The bound is tighter than that recently obtained by Kashyap Both indicate that with ideal distributed lossy coding, dense sensor networks can efficiently sense and convey a field, in contrast to the negative result obtained by Marco for encoders based on scalar quantization and Slepian-Wolf distributed lossless coding. The second scheme is transform coding with scalar coefficient quantization. A new generalized transform coding analysis, as well as the aforementioned convergence theorems, is used to find the smallest attainable rate at asymptotically large sampling rates in terms of the source spectral density and the operational rate-distortion function of the family of quantizers, which in contrast to previous analyses need not be convex. The result shows that when a transform is used, scalar quantization need not cause the poor performance found by Marco As a corollary, the final result pursues an approach, originally proposed by Berger, to show that the inverse water-pouring formula for the rate-distortion function can be attained at high sampling rates by transform coding with ideal vector quantization to encode the coefficients. Also established in the paper are relations between operational rate-distortion and distortion-rate functions for a continuous parameter source and those for the discrete parameter source that results from sampling.

MAP Detection and Robust Lossy Coding Over Soft-Decision Correlated Fading Channels

The joint source-channel coding (JSCC) problem for soft-decision-demodulated time-correlated fading channels is investigated without the use of channel coding and interleaving. For the purpose of system design, the recently introduced nonbinary noise discrete channel with queue-based (QB) noise (NBNDC-QB) is adopted. This analytically tractable Markovian model has been shown to represent effectively correlated fading channels that are hard to handle analytically. Optimal sequence maximum a posteriori (MAP) detection of a discrete Markov source sent over the NBNDC-QB is first studied. When the Markov source is binary and symmetric, a necessary and sufficient condition under which the MAP decoder is reduced to a simple instantaneous symbol-by-symbol decoder is established. Two robust lossy source coding schemes with a low encoding delay are next proposed for the NBNDC-QB. The first scheme consists of a scalar quantizer (SQ), a proper index assignment, and a sequence MAP decoder designed to harness the redundancy left in the indexes of the quantizer and the soft-decision output and the noise correlation of the channel. The second scheme is a classical noise-resilient vector quantizer known as the channel-optimized vector quantizer (COVQ). It is demonstrated that both systems can successfully exploit the memory and soft-decision information of the channel. Signal-to-distortion-ratio (SDR) gains of more than 1.7 dB are obtained over hard-decision demodulation by using only two bits for soft decision. Furthermore, gains as high as 4.4 dB can be achieved for a strongly correlated channel, in comparison with systems designed for the ideally interleaved (memoryless) channel. Finally, it is numerically observed that, for low coding rates, the NBNDC-QB model can accurately approximate discrete fading channels (DFCs) in terms of SDR performance.

Compute-and-Forward Strategies for Cooperative Distributed Antenna Systems

We study a distributed antenna system where L antenna terminals (ATs) are connected to a central processor (CP) via digital error-free links of finite capacity R0, and serve K user terminals (UTs). This model has been widely investigated both for the uplink (UTs to CP) and for the downlink (CP to UTs), which are instances of the general multiple-access relay and broadcast relay networks. We contribute to the subject in the following ways: 1) For the uplink, we consider the recently proposed “compute and forward” (CoF) approach and examine the corresponding system optimization at finite SNR. 2) For the downlink, we propose a novel precoding scheme nicknamed “reverse compute and forward” (RCoF). 3) In both cases, we present low-complexity versions of CoF and RCoF based on standard scalar quantization at the receivers, that lead to discrete-input discrete-output symmetric memoryless channel models for which near-optimal performance can be achieved by standard single-user linear coding. 4) We provide extensive numerical results and finite SNR comparison with other “state of the art” information theoretic techniques, in scenarios including fading and shadowing. The proposed uplink and downlink system optimization focuses specifically on the ATs and UTs selection problem. In both cases, for a given set of transmitters, the goal consists of selecting a subset of the receivers such that the corresponding system matrix has full rank and the sum rate is maximized. We present low-complexity ATs and UTs selection schemes and demonstrate through Monte Carlo simulation that the proposed schemes essentially eliminate the problem of rank deficiency of the system matrix and greatly mitigate the noninteger penalty affecting CoF/RCoF at high SNR. Comparison with other state-of-the art information theoretic schemes, show competitive performance of the proposed approaches with significantly lower complexity.

Block Adaptive Scalar-Vector Quantization Used to Compress SAR Raw Data

A block adaptive scalar-vector quantization (BASVQ) algorithm is put forward in this paper, namely, scalar quantization is applied when data blocks satisfy Gaussian distribution while vector quantization is applied when doesnt satisfy. The experiments demonstrate that the performance of BASVQ algorithm surpasses that of BAQ algorithm. The BASVQ algorithm has practical value in some degree.

Joint Source-Channel Coding for Deep-Space Image Transmission using Rateless Codes

A new scheme for image transmission over noisy channel is proposed. Similar to standard image compression, the scheme includes a transform followed by successive refinement scalar quantization. Unlike conventional schemes, though, in the proposed system the quantized transform coefficients are linearly mapped into channel symbols using systematic encoders. This fixed-to-fixed length linear index coding approach avoids the use of an explicit entropy stage (e.g., arithmetic or Huffman coding), which is typically fragile to channel post-decoding residual errors. We use codes over GF(4), which are particularly suited for this application, since they are matched to the dead-zone quantizer symbol alphabet and to the QPSK modulation used on the deep-space communication channel. We optimize the proposed system where the codes are systematic Raptor codes over GF(4). The rateless property of Raptor encoders allows to implement a continuum of rates, in order to accurately match the channel rate to the transmission channel capacity and to the quantized source entropy rate for each transform subband and refinement level. Comparisons are provided with respect to the concatenation of state-of-the-art image and channel schemes used by Jet Propulsion Laboratories (JPL) for the Mars Exploration Rover (MER) Mission.

Self-organizing maps for the design of multiple description vector quantizers

Multiple description coding is an appealing tool to guarantee graceful signal degradation in the presence of unreliable channels. While the principles of multiple description scalar quantization are well-understood and solid guidelines exist to design effective systems, the same does not hold for vector quantization, especially at low bit-rates, where burdensome and unsatisfactory design techniques discourage its use altogether in applications.In this work we use the self-organizing maps to design multiple description VQ codebooks. The proposed algorithm is flexible, fast and effective: it deals easily with a large variety of situations, including the case of more than two descriptions, with a computational complexity that remains fully affordable even for large codebooks, and a performance comparable to that of reference techniques.A thorough experimental analysis, conducted in a wide range of operating conditions, proves the proposed technique to perform on par with well-known reference methods based on greedy optimization, but with a much lower computational burden. In addition, the resulting codebook can be itself optimized, thus providing even better performance. All experiments are fully reproducible, with all software and data available online for the interested researchers.

An Efficient Hybrid Image Coding Scheme Combining Wavelets, Neural Networks and Differential Pulse Code Modulation for Effectual Image Compression

Large images consume more storage space needing high data rates for transmission demanding the innovation of efficient image compression systems. Owing to the massive parallel architecture and generalization ability of neural networks to memorize inputs even on untrained data, the computational simplicity of wavelets, ability of Differential Pulse Code Modulation (DPCM) to reduce the unused or redundant bits in the information, in this paper an hybrid image compression system combining the advantages of wavelets and neural networks is implemented along with Differential Pulse Code Modulation based on the predicted sample values. Scalar quantization and Huffman encoding schemes are used as well for compressing different sub bands i.e the low frequency band coefficients are compressed by the DPCM while the high frequency band coefficients are compressed using neural networks. Satisfactory reconstructed images with increased bit rates and large Peak Signal to Noise Ratio (PSNR) can be achieved with this scheme. Wavelet transform eliminates the blocking artefacts’ associated with cosine transform and neural networks minimize the Mean Square Error (MSE).Empirical analysis and metrics calculation is performed for the sake of relative analysis. General Terms Artificial Neural Networks, Fidelity, Mean Square Error, Peak Signal to Noise Ratio, Quantization and Huffman encoding.