Classified Vector Quantization Research Articles

Image coding based on classified vector quantisation using edge orientation patterns

Vector quantisation (VQ) shows a good performance for image coding with high-compression ratios. However, there are many difficulties for image coding with VQ, especially the edge degradation and high-computational complexity. To resolve these two problems, the authors propose a new coding method based on edge orientation patterns (EOPs) by classifying image blocks into nine classes according to their edge orientations. For colour image coding, 27 codebooks (nine for each colour component) are pre-designed based on a series training images. In the encoding stage, an input colour image is decomposed into Y, Cb, and Cr components, and each component image is divided into non-overlapping 4 × 4 blocks. For each block, eight edge orientation templates of size 4 × 4 are performed to determine its edge orientation. According to the edge orientation, each block is compressed by using the corresponding codebook. Essentially, the authors’ scheme is a kind of classified VC (CVQ). Simulation results show that, their EOP-based CVQ can largely improve the compression efficiency as well as speeding up the encoding process and it is sufficient to establish effectiveness of the authors’ algorithm as compared with the existing techniques.

Text-dependent speaker verification using classical LBG, adaptive LBG and FCM vector quantization

An important task of speaker verification is to generate speaker specific models and match an input speaker’s utterance with these models. This paper focuses on comparing the performance of text dependent speaker verification system using Mel Frequency Cepstral Coefficients feature and different Vector Quantization (VQ) based speaker modelling techniques to generate the speaker specific models. Speaker-specific information is mainly represented by spectral features and using these features we have developed the model which serves as an important entity for determining the claimed identity of the speaker. In the modelling part, we used Linde, Buzo, Gray (LBG) VQ, proposed adaptive LBG VQ and Fuzzy C Means (FCM) VQ for generating speaker specific model. The experimental results that are performed on microphonic database shows that accuracy significantly depends on the size of the codebook in all VQ techniques, and on FCM VQ accuracy also depend on the value of learning parameter of the objective function. Experiment results shows that how the accuracy of speaker verification system is depend on different representations of the codebook, different size of codebook in VQ modelling techniques and learning parameter in FCM VQ.

Classified Vector Quantization and its Application on Compression of Iris Images in the Safety of Marine Systems

As a standard, 512 byte IrisCode templates developed with specific algorithms are stored in databases and used in iris recognition process. Future tendencies are to use exclusively real iris images rather than IrisCode templates in the iris recognition process. Many of current iris recognition systems use portable devices (e.g. iris scanners) which are often required to transmit image or template over communication channel. Image compression can be used in order to reduce the transmission time and storage capacities. Classified Vector Quantization (CVQ) and ordinary Vector Quantization (VQ) are used for compression of greyscale iris images collected from one of the available public databases of iris images. Results show that both compression methods are significantly more effective when applied to iris images than when applied to average images from everyday environments since iris images are fairly uniform and contain lowcontrast levels. Originally, CVQ is used to improve the quality of edges of compressed images because they are the most important part of image for visual impression on humans. The paper presents the comparison and major advantage of CVQ over ordinary VQ in terms of significant time reduction needed for iris images to be coded, and therefore it highlights a new important application of CVQ.

Prediction Transform GMM Vector Quantization for Wideband LSFs DOI: 10.14209/jcis.2014.5

Classified vector quantization (VQ) using a Gaussian Mixture Model (GMM) preclassifier is a powerful VQ method. A prediction-based lower-triangular transform (PLT) is proposed for the enhancement of VQ in each cluster. The PLT is defined for generic vector spaces in the context of the covariance method of linear prediction (LP). Optimal quantizer banks are designed in minimum noise structures whose codebooks are used for the proposed Cartesian split VQ (CSVQ), which improves their coding gain. CSVQ is tested forline spectral frequency (LSF) quantization of wideband speech spectra, revealing a comparable average performance to the Karhunen-Loeve transform (KLT) at lower rates with reduced outlier generation and computational complexity.

Prediction Transform GMM Vector Quantization for Wideband LSFs

Classified vector quantization (VQ) using a Gaussian Mixture Model (GMM) preclassifier is a powerful VQ method. A prediction-based lower-triangular transform (PLT) is proposed for the enhancement of VQ in each cluster. The PLT is defined for generic vector spaces in the context of the covariance method of linear prediction (LP). Optimal quantizer banks are designed in minimum noise structures whose codebooks are used for the proposed Cartesian split VQ (CSVQ), which improves their coding gain. CSVQ is tested forline spectral frequency (LSF) quantization of wideband speech spectra, revealing a comparable average performance to the Karhunen-Loeve transform (KLT) at lower rates with reduced outlier generation and computational complexity.

Super-resolution algorithm based on weighted vector quantization

Super resolution is a technique that obtains high-resolution (HR) images from the corresponding low-resolution (LR) ones and moreover, makes these HR images as natural as possible. There are usually some artifacts of these HR images caused by the lack of the information of the detailed textures (high-frequency information). In our research, an algorithm based on weighted vector quantization is proposed to predict the high-frequency information. First of all, the classified vector quantization is adopted to establish a pair of codebooks of high-frequency information for LR and HR patches. Second, three code vectors are selected from the LR codebooks based on the correlation coefficients to establish a multilinear regression model with the input LR patch and meanwhile the weights are decided adaptively. Then, three corresponding HR code vectors are used with the weights to reconstruct the high-frequency information of the HR patch. Finally, the reconstructed high-frequency information is used to refine the preliminarily upscaled HR image. The experimental results show that the performance of the proposed algorithm is good not only in objective but also in subjective measurement.

Image retrieval based on quadtree classified vector quantization

In this paper, a color image retrieval scheme based on quadtree classified vector quantization (QCVQ) is proposed. This scheme not only captures intra-block correlation but also exploits the visual importance of image blocks to efficiently describe the content of images in a compressed domain. In the proposed algorithm, a query image is first divided by quadtree segmentation and then classified into smooth and high-detail blocks. For high-detail blocks, the local thresholding classifier with 28 edge binary templates is employed to extract a variety of visually important regions which are edge intensive. After all of the blocks in the image are encoded by the pre-trained QCVQ codebook, the indices in the compressed domain are obtained. Finally, the frequencies of indices are counted to build the index histogram as a feature of the query image. Simulation results demonstrate that our proposed scheme yields the better retrieval performance compared to the well-known vector quantization (VQ)-based image retrieval method and three other techniques. These results show that quadtree segmentation and edge style classification are indeed helpful for improving the performance of content-based image retrieval.

Vehicle Classification with Confidence by Classified Vector Quantization

Automated vehicle classification based on static images is highly practical and directly applicable for various operations such as traffic related investigations. An integrated vehicle detection and classification system is proposed in this paper. A multi-resolution vehicle detection scheme is introduced as an improvement over the cascade boosted classifiers proposed recently by Negri et al. 2008 in the literature. Building on solutions from previous works from Negri et al, the implementation of a new decision strategy renders current detection method to be robust to environmental changes. The vehicle classification is based on the Classified Vector Quantization (CVQ) proposed earlier by Zhang et al. 2009. The justification of choosing CVQ is its advantages in providing classification confidence by incorporating rejection option. The significance of rejection in enhancing the system?s reliability is emphasized and evaluated. A database composed of more than 2800 images of four types of vehicles (cars, vans, light trucks and buses) was created using police surveillance cameras. The proposed scheme offers a performance accuracy of over 95% with a rejection rate of 8%, and reliability over 98% with a rejection rate of 20%. This exhibits promising potentials for implementations into real-world applications.

Essentials of the self-organizing map

The self-organizing map (SOM) is an automatic data-analysis method. It is widely applied to clustering problems and data exploration in industry, finance, natural sciences, and linguistics. The most extensive applications, exemplified in this paper, can be found in the management of massive textual databases and in bioinformatics. The SOM is related to the classical vector quantization (VQ), which is used extensively in digital signal processing and transmission. Like in VQ, the SOM represents a distribution of input data items using a finite set of models. In the SOM, however, these models are automatically associated with the nodes of a regular (usually two-dimensional) grid in an orderly fashion such that more similar models become automatically associated with nodes that are adjacent in the grid, whereas less similar models are situated farther away from each other in the grid. This organization, a kind of similarity diagram of the models, makes it possible to obtain an insight into the topographic relationships of data, especially of high-dimensional data items. If the data items belong to certain predetermined classes, the models (and the nodes) can be calibrated according to these classes. An unknown input item is then classified according to that node, the model of which is most similar with it in some metric used in the construction of the SOM. A new finding introduced in this paper is that an input item can even more accurately be represented by a linear mixture of a few best-matching models. This becomes possible by a least-squares fitting procedure where the coefficients in the linear mixture of models are constrained to nonnegative values.

A Novel Approach Based on CVQ Technique for Face Recognition

Face recognition has been widely used to identify faces according to the patterns of mouth, lip, eyes and so on. Existing face recognition techniques provide a reasonable accuracy. However, their effectiveness significantly degrades in presence of uncontrolled conditions such as variations of background light and image sizes. To deal with this problem, we propose a novel approach based on Classified Vector Quantization (CVQ) technique for face recognition. The new approach divides images into some blocks and each block is classified into several patterns. The Vector Quantization technique (VQ) is applied on the vectors of each pattern. A major benefit of this technique is to consider the geometric structure of face and skin correlations which provide recognition based on changes on human's age. In order to evaluate our approach, experiments on some standard image databases have been conducted. The results demonstrate the high recognition rate and accuracy of our technique.

Scene Categorization with Classified Codebook Model

This paper presents an efficient yet powerful codebook model, named classified codebook model, to categorize natural scene category. The current codebook model typically resorts to large codebook to obtain higher performance for scene categorization, which severely limits the practical applicability of the model. Our model formulates the codebook model with the theory of vector quantization, and thus uses the famous technique of classified vector quantization for scene-category modeling. The significant feature in our model is that it is beneficial for scene categorization, especially at small codebook size, while saving much computation complexity for quantization. We evaluate the proposed model on a well-known challenging scene dataset: 15 Natural Scenes. The experiments have demonstrated that our model can decrease the computation time for codebook generation. What is more, our model can get better performance for scene categorization, and the gain of performance becomes more pronounced at small codebook size.

KLT-based adaptive entropy-constrained quantization with universal arithmetic coding

For flexible speech coding, a Karhunen-Loève Transform (KLT) based adaptive entropy-constrained quantization (KLT-AECQ) method is proposed. It is composed of backward-adaptive linear predictive coding (LPC) estimation, KLT estimation based on the time-varying LPC coefficients, scalar quantization of the speech signal in a KLT domain, and superframe-based universal arithmetic coding based on the estimated KLT statistics. To minimize the outliers both in rate and distortion, a new distortion criterion includes the penalty in the rate increase. Gain adaptive step size selection and bounded Gaussian source model also cooperate to increase the perceptual quality. KLT-AECQ does not require either any explicit codebook or a training step, thus KLT-AECQ can have an infinite number of rate-distortion operating points regardless of time-varying source statistics. For the speech signal, the conventional KLT-based classified vector quantization (KLT-CVQ) and the proposed KLT-AECQ yield signal-to-noise ratios of 17.86 and 26.22, respectively, at around 16 kbits/s. The perceptual evaluation of speech quality (PESQ) scores for each method are 3.87 and 4.04, respectively <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">.</sup>

KLT-based adaptive entropy-constrained vector quantization for the speech signals

For efficient variable-rate speech coding, Karhunen-Loeve transform based adaptive entropy-constrained vector quantization (KLT-AECVQ) is proposed. The proposed method consists of backward-adaptive linear predictive coding (LPC) analysis, KLT estimation based on LPC coefficients, and lattice vector quantization followed by Huffman coding according to KLT statistics. As different statistics in an original-signal domain can be mapped into identical statistics in a KLT domain, only a few classified Huffman codebooks are sufficient to represent KLT-domain source statistics. KLT-AECVQ with 32 Huffman codebooks has comparable rate-distortion performance with theoretically optimal AECVQ with infinite number of Huffman codebooks. KLT-AECVQ also produces superior perceptual quality to KLT-based classified vector quantization (KLTCVQ) that yielded better quality than conventional code excited linear predictive (CELP) codec. Under five-sample delay constraints, KLT-AECVQ has also three times lower complexity than CELP codec.

Novel hybrid classified vector quantization using discrete cosine transform for image compression

We present a novel image compression technique using a classified vector Quantizer and singular value decomposition for the efficient representation of still images. The proposed method is called hybrid classified vector quantization. It involves a simple but efficient classifier-based gradient method in the spatial domain, which employs only one threshold to determine the class of the input image block, and uses three AC coefficients of discrete cosine transform coefficients to determine the orientation of the block without employing any threshold. The proposed technique is benchmarked with each of the standard vector quantizers generated using the k-means algorithm, standard classified vector quantizer schemes, and JPEG-2000. Simulation results indicate that the proposed approach alleviates edge degradation and can reconstruct good visual quality images with higher peak signal-to-noise ratio than the benchmarked techniques, or be competitive with them.

Classified Vector Quantisation and population decoding for pattern recognition

Learning Vector Quantisation (LVQ) is a method of applying the Vector Quantisation (VQ) to generate references for Nearest Neighbour (NN) classification. Though successful in many occasions, LVQ suffers from several shortcomings, especially the reference vectors are prone to diverge. In this paper, we propose a Classified Vector Quantisation (CVQ) to establish VQ for classification. By CVQ, each data category is represented by its own codebook, which can be implemented by some learning algorithms. In classification process, each codebook offers a generalised NN. The examples of handwritten digit recognition and offline signature verification are used to demonstrate the efficiency of the proposed scheme.

Image compression using improved hybrid classified vector quantisation

A novel image compression technique using classified vector quantiser and singular value decomposition is proposed for the efficient representation of still images. The proposed method is called improved hybrid classified vector quantisation. The proposed technique was benchmarked with the standard vector quantiser generated using the k-means algorithm, and JPEG-2000. Simulation results indicate that the proposed approach alleviates edge degradation and can reconstruct good visual quality images with higher peak signal-to-noise ratio than the benchmarked techniques.

On the encoding rate and discrete modulation adaptation design for MIMO links

It is well-known that MIMO links in slow fading channels can benefit significantly through rate and power adaptations when there is knowledge of channel state information at the transmitter (CSIT). However, practical MIMO transmitter can only support finite combinations of the channel encoding rates and modulation constellation levels due to the finite number of bits allocated in the control overhead (embedded in the packet transmission). In this paper, we shall address various practical MIMO adaptation design issues associated with discrete transmission modes. We propose a systematic design framework to address the MIMO adaptation design based on information theoretical framework. The design problem is cast into an optimization problem similar to the classical vector quantization problem with a modified distortion measure. As an illustration, we apply the design framework to a 2 times 2 MIMO link and compare the goodput performance with the optimal performance predicted by the theory. We found that the goodput performance of the MIMO link predicted by the theoretical framework matches the actual performance closely

Improvement of interpolated color filter array image using modified mean-removed classified vector quantization

To reduce the cost of digital cameras, the color filter array (CFA) is usually coated upon a single-chip image sensor. Each pixel of the image sensor can sense only one of the R, G, B colors under CFA. The two missing colors of a pixel are estimated using the CFA interpolation algorithm. The interpolation algorithm may generate color distortion and degrade the quality of interpolated images. To improve the quality of interpolated images, a modified mean-removed classified vector quantization algorithm is proposed to reduce the estimation error of interpolated colors. The algorithm extends and modifies vector quantization to discover the relationships between the interpolated images and their corresponding original versions using the information from CFA images. The discovered relationships are stored in codebooks and are used to improve the quality of images interpolated by the existing CFA interpolation algorithms. The experiments reveal that the proposed algorithm can improve the quality of images interpolated by the best CFA interpolation algorithm as far as we know. In terms of PSNR, the average PSNR improvements of R, G, and B channels are 0.89, 0.71, and 0.74 dB, respectively.

Artifact reduction of compressed color images using modified mean‐removed classified vector quantization

In this paper, a new artifact reduction algorithm for compressed color images using MMRCVQ is proposed. The algorithm extends and modifies vector quantization (VQ) for discovering the relationships between the uncompressed color images and their deblocked compressed versions by classifying the deblocked compressed blocks into several categories using information from their neighboring blocks. The discovered relationships are stored in two codebooks and are used to recover the missing information of compressed color images. To increase the availability of codewords and reduce the memory needed for storing codewords, mean‐removed vectors are used to generate codebooks. The experimental results show that the proposed approach can remove, effectively, the artifacts caused by high compression and improve perceptual quality significantly. Compared to existing methods, the proposed approach usually uses much less computing time to recover a compressed color image and has much better image quality.

KLT-Based Adaptive Classified VQ of the Speech Signal

Compared to scalar quantization (SQ), vector quantization (VQ) has memory, space-filling, and shape advantages. If the signal statistics are known, direct vector quantization (DVQ) according to these statistics provides the highest coding efficiency, but requires unmanageable storage requirements if the statistics are time varying. In code-excited linear predictive (CELP) coding, a single compromise codebook is trained in the excitation-domain and the space-filling and shape advantages of VQ are utilized in a nonoptimal, average sense. In this paper, we propose Karhunen-Loe/spl grave/ve transform (KLT)-based adaptive classified VQ (CVQ), where the space-filling advantage can be utilized since the Voronoi-region shape is not affected by the KLT. The memory and shape advantages can be also used, since each codebook is designed based on a narrow class of KLT-domain statistics. We further improve basic KLT-CVQ with companding. The companding utilizes the shape advantage of VQ more efficiently. Our experiments show that KLT-CVQ provides a higher SNR than basic CELP coding, and has a computational complexity similar to DVQ and much lower than CELP. With companding, even single-class KLT-CVQ outperforms CELP, both in terms of SNR and codebook search complexity.