Vector Quantization Technique Research Articles

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.

Image indexing and retrieval in JPEG compressed domain based on vector quantization

This paper presents a new and effective image indexing technique that extracts features from JPEG compressed images. Using vector quantization techniques (VQ) and a codebook generated using a K-means clustering algorithm, the proposed technique is able to create an effective histogram from DCT coefficients, which are the major components of JPEG. The proposed method only needs to do partial decoding, therefore it can accelerate the work of indexing images. Experimental results show that its retrieval performance is higher compared with those of other methods from pixel and compressed domains. All experiments confirm the effectiveness and efficiency of the proposed method.

Real-Time Hand Gesture Detection and Recognition Using Bag-of-Features and Support Vector Machine Techniques

This paper presents a novel and real-time system for interaction with an application or video game via hand gestures. Our system includes detecting and tracking bare hand in cluttered background using skin detection and hand posture contour comparison algorithm after face subtraction, recognizing hand gestures via bag-of-features and multiclass support vector machine (SVM) and building a grammar that generates gesture commands to control an application. In the training stage, after extracting the keypoints for every training image using the scale invariance feature transform (SIFT), a vector quantization technique will map keypoints from every training image into a unified dimensional histogram vector (bag-of-words) after K-means clustering. This histogram is treated as an input vector for a multiclass SVM to build the training classifier. In the testing stage, for every frame captured from a webcam, the hand is detected using our algorithm, then, the keypoints are extracted for every small image that contains the detected hand gesture only and fed into the cluster model to map them into a bag-of-words vector, which is finally fed into the multiclass SVM training classifier to recognize the hand gesture.

A high quality and small shadow size visual secret sharing scheme based on hybrid strategy for grayscale images

Many secret sharing schemes for digital images have been developed in recent decades. Traditional schemes typically must deal with the problem of computational complexity, and other visual secret sharing schemes come with a higher transmission cost and storage cost; that is, each shadow size is m times as big as the original secret image. The new ( 2 , n ) secret sharing scheme for grayscale images proposed in this paper is based a combination of acceptable image quality using block truncation coding (BTC), high compression ratio discrete wavelet transform (DWT) and good subjective performance of the vector quantization (VQ) technique. Experimental results confirm that our proposed scheme not only generates a high quality reconstructed original image but also generates small, random-like grayscale shadows.

An algorithm for binary codebook design based on the average bitmap replacement error (ABPRE)

In this paper, an algorithm for binary codebook design has been used in vector quantization technique, which is used to improve the acceptability of the absolute moment block truncation coding (AMBTC) method. Vector quantization (VQ) method is used to compress the bitmap (the output proposed from the first method (AMBTC)). In this paper, the binary codebook can be engender for many images depending on randomly chosen to the code vectors from a set of binary images vectors, and this codebook is then used to compress all bitmaps of these images. The chosen of the bitmap of image in order to compress it by using this codebook based on the criterion of the average bitmap replacement error (ABPRE). This paper is suitable to reduce bit rates (increase compression ratios) with little reduction of performance (PSNR).

An algorithm for binary codebook design based on the average bitmap replacement error (ABPRE)

In this paper, an algorithm for binary codebook design has been used in vector quantization technique, which is used to improve the acceptability of the absolute moment block truncation coding (AMBTC) method. Vector quantization (VQ) method is used to compress the bitmap (the output proposed from the first method (AMBTC)). In this paper, the binary codebook can be engender for many images depending on randomly chosen to the code vectors from a set of binary images vectors, and this codebook is then used to compress all bitmaps of these images. The chosen of the bitmap of image in order to compress it by using this codebook based on the criterion of the average bitmap replacement error (ABPRE). This paper is suitable to reduce bit rates (increase compression ratios) with little reduction of performance (PSNR).

Combining LVQ with SVM technique for image semantic annotation

When support vector machine (SVM) classifier is applied to image semantic annotation, it usually encounters the problem of excessive training samples. In this paper, we propose a novel method, which is by combining learning vector quantization (LVQ) technique and SVM classifier, to improve annotation accuracy and speed. Affinity propagation algorithm-based LVQ technique is used to optimize the training set, and a few number of optimized representative feature vectors are used to train SVM. This approach not only meets the small sample size characteristic of SVM, but also greatly accelerates the training and annotating process. Comparative experimental studies confirm the validity of the proposed method.

MareText Independent Speaker Identification based on K-mean Algorithm

This paper proposes a text-independent speaker identification system based on Mel Frequency Cepstral Coefficients as a feature extraction and Vector Quantization technique that would minimize the data required for processing. The correlation between the identification success rate and the various parameters of the system including the feature extraction tools and the data minimization technique will be examined. Extracted features of a speaker are quantized by a number of centroids and the K-mean algorithm has been integrated into the proposed speaker identification system. Such centroids constitute the codebook of that speaker. MFCC are calculated in both training and testing phases. To calculate these MFCC speakers uttered different words, once in a training session and once in a testing one. The speakers were identified according to the minimum quantization distance which was calculated between the centroids of each speaker in the training phase and the MFCC of individual speakers in the testing phase. Analysis was carried out to identify parameter values that could be used to improve the performance of the system. The experimental results illustrate the efficiency of the proposed method under several conditions Speaker recognition aims at recognizing speakers from their voices as each person has his own speech characteristics and his way of speaking. Speaker recognition is basically divided into speaker identification and speaker verification. Speaker identification is the process of determining which registered speaker provides the speech input, while verification is the task of automatically determining if a person really is the person he or she claims to be. Speaker recognition has many particular applications as a speaker's voice can be used to verify their identity and control access to services such as banking by telephone, database access services, voice dialing telephone shopping, information services and voice mail. Another important application of speaker recognition technology is for forensic purposes (1). Speaker recognition can be classified as based on text-dependent or text-independent methods. In the text dependent method, the speaker has to say key words or sentences having the same text for both training and recognition trials. Whereas in the text independent case the system can identify the speaker regardless of what is being said (2), (3), (4). The goal of this study is a real time text-independent speaker identification system, which consists of comparing a speech signal from an unknown speaker to a database of known speakers. The system will operate in two modes: a training mode and a recognition mode. During the training mode users will record their voices and make a feature model it. The recognition mode will use the information that the user has provided in the training mode and attempt to isolate and identify the speaker. The Mel Frequency Cepstral Coefficients (MFCC) and the Vector Quantization (VQ) algorithms are used to implement this process. The simple K-means clustering algorithm is used in this study whereas the LBG is used in other similar work (4).

Building compact recognizers of handwritten Chinese characters using precision constrained Gaussian model, minimum classification error training and parameter compression

In our previous work, a so-called precision constrained Gaussian model (PCGM) was proposed for character modeling to design compact recognizers of handwritten Chinese characters. A maximum likelihood training procedure was developed to estimate model parameters from training data. In this paper, we extend the above-mentioned work by using minimum classification error (MCE) training to improve recognition accuracy and using both split vector quantization and scalar quantization techniques to further compress model parameters. Experimental results on a handwritten character recognition task with a vocabulary of 2,965 Kanji characters demonstrate that MCE-trained and compressed PCGM-based classifiers can achieve much higher recognition accuracies than their counterparts based on traditional modified quadratic discriminant function (MQDF) when the footprint of the classifiers has to be made very small, e.g., less than 2 MB.

Frame permutation quantization

Frame permutation quantization (FPQ) is a new vector quantization technique using finite frames. In FPQ, a vector is encoded using a permutation source code to quantize its frame expansion. This means that the encoding is a partial ordering of the frame expansion coefficients. Compared to ordinary permutation source coding, FPQ produces a greater number of possible quantization rates and a higher maximum rate. Various representations for the partitions induced by FPQ are presented, and reconstruction algorithms based on linear programming, quadratic programming, and recursive orthogonal projection are derived. Implementations of the linear and quadratic programming algorithms for uniform and Gaussian sources show performance improvements over entropy-constrained scalar quantization for certain combinations of vector dimension and coding rate. Monte Carlo evaluation of the recursive algorithm shows that mean-squared error (MSE) decays as M − 4 for an M-element frame, which is consistent with previous results on optimal decay of MSE. Reconstruction using the canonical dual frame is also studied, and several results relate properties of the analysis frame to whether linear reconstruction techniques provide consistent reconstructions.

The analysis of motor vehicle crash clusters using the vector quantization technique

AbstractIn this paper, a powerful tool for analyzing motor vehicle data based on the vector quantization (VQ) technique is demonstrated. The technique uses an approximation of a probability density function for a stochastic vector without assuming an “a priori” distribution. A self‐organizing map (SOM) is used to transform accident data from an N‐dimensional space into a two‐dimensional plane. The SOM retains all the original data yet provides an effective visual tool for describing patterns such as the frequency at which a particular category of events occurs. This enables new relationships to be identified. Accident data from three cities in Italy (Turin, Milan, and Legnano) are used to illustrate the usefulness of the technique. Crashes are aggregated and clustered crashes by type, severity, and along other dimensions. The paper includes discussion as to how this method can be utilized to further improve safety analysis. Copyright © 2010 John Wiley & Sons, Ltd.

VQSVM: A case study for incorporating prior domain knowledge into inductive machine learning

When dealing with real-world problems, there is considerable amount of prior domain knowledge that can provide insights on various aspect of the problem. On the other hand, many machine learning methods rely solely on the data sets for their learning phase and do not take into account any explicitly expressed domain knowledge. This paper proposes a framework that investigates and enables the incorporation of prior domain knowledge with respect to three key characteristics of inductive machine learning algorithms: consistency, generalization and convergence. The framework is used to review, classify and analyse key existing approaches to incorporating domain knowledge into inductive machine learning, as well as to consider the risks of doing so. The paper also demonstrates the design of a novel hierarchical semi-parametric machine learning method, capable of incorporating prior domain knowledge. The method—VQSVM—extends the support vector machine (SVM) family of methods with vector quantization (VQ) techniques to address the problem of learning from imbalanced data sets. The paper presents the results of testing the method on a collection of imbalanced data sets with various imbalance ratios and various numbers of subclasses. The learning process of the VQSVM method utilizes some domain knowledge to solve problem of fitting imbalance data. The experiments in the paper demonstrate that enabling the incorporation of prior domain knowledge into the SVM framework is an effective way to overcome the sensitivity of SVM towards the imbalance ratio in a data set.

Tumor Demarcation in Mammography Images using LBG on Probability Image

The ability to improve diagnostic information from medical images can be enhanced by designing computer processing algorithms that is why we proposed new algorithm to detect cancer in mammogram breast cancer images. In this paper we proposed segmentation using vector quantization technique. Here we used Linde Buzo and Gray (LBG)for segmentation of mammographic images on probability image. Initially probability of input image is calculated and displayed as a result. In second step a codebook of size 128 was generated for probability image. These code vectors were further reclustered in 8 clusters using same LBG algorithm. These 8 images were displayed as a result. This approach does not leads to over segmentation or under segmentation. For the comparison purpose we displayed results of GLCM and watershed segmentation along with this method.

An optimized unitary beamforming technique for MIMO broadcast channels

This paper addresses the problem of linear beamforming design in MIMO broadcast channels. An iterative optimization method for unitary beamforming is proposed, based on successive optimization of Givens rotations. Under the assumption of perfect channel state information at the transmitter (CSIT) and for practical average signal-to-noise ratios (SNR), the proposed technique provides higher sum rates than zero-forcing (ZF) beamforming while performing close to minimum-mean- squared-error (MMSE) beamforming when the number of transmit antennas equals the number of scheduled users. Moreover, it is shown to achieve linear sum-rate growth with the number of transmit antennas. Interestingly, the proposed unitary beamforming approach proves to be very robust to channel estimation errors. In the simulated scenarios, it provides better sum rates than ZF beamforming and even MMSE beamforming as the variance of the estimation error increases. When combined with simple vector quantization techniques for CSIT feedback in systems with multiuser scheduling, the proposed technique proves to be well suited for limited feedback scenarios with practical number of users, exhibiting performance gains over existing techniques.

Vector Quantization Techniques For Partial Encryption of Wavelet-based Compressed Digital Images

Vector Quantization Techniques For Partial Encryption of Wavelet-based Compressed Digital Images

Deterministic vector long-term forecasting for fuzzy time series

In the last decade, fuzzy time series have received more attention due their ability to deal with the vagueness and incompleteness inherent in time series data. Although various improvements, such as high-order models, have been developed to enhance the forecasting performance of fuzzy time series, their forecasting capability is mostly limited to short-term time spans and the forecasting of a single future value in one step. This paper presents a new method to overcome this shortcoming, called deterministic vector long-term forecasting (DVL). The proposed method, built on the basis of our previous deterministic forecasting method that does not require the overhead of determining the order number, as in other high-order models, utilizes a vector quantization technique to support forecasting if there are no matching historical patterns, which is usually the case with long-term forecasting. The vector forecasting method is further realized by seamlessly integrating it with the sliding window scheme. Finally, the forecasting effectiveness and stability of DVL are validated and compared by performing Monte Carlo simulations on real-world data sets.

Lip Vermilion Profile Patterns and Corresponding Dentoskeletal Forms in Female Adults

To objectively classify shapes of the human lip vermilion in the lateral view, and to examine whether any morphologic characteristics of dentoskeletal patterns are specific to each classified lip profile pattern. Pretreatment, lateral facial photographs of 234 Japanese women were selected. Investigators with expert knowledge of the anatomic traits of the lip vermilion in the lateral view extracted from images of the face a 13-dimensional feature vector that represented lip vermilion profile shapes. The vector quantization technique was applied to the feature vectors to mathematically optimize the number of lip vermilion profile patterns. Dentoskeletal patterns that corresponded to each classified lip shape were compared statistically. Seven patterns were found, and differences between patterns were notably maximized by the flatness of the anterior portion of the lip vermilion for the upper and lower lip, the position of the most protruded point of the lip vermilion, lip fissure inclination, and differences between the horizontal positions of the upper and lower lip vermilions. The dentoskeletal forms showed significant differences between classified lip vermilion profile patterns (P < .01). (1) Vector quantization revealed that classifying lip vermilion profiles into seven representative patterns was optimal for maximizing differences in the configuration of the lip vermilion. (2) Lip vermilion profile shapes were found to be associated with horizontal lengths of the anterior cranial base, horizontal/vertical positions, inclination and length of the mandible, and horizontal positions and labio-lingual inclinations of the upper and lower incisors.

Visually imperceptible image hiding scheme based on vector quantization

A novel visually imperceptible image hiding scheme based on VQ compression and LSB modification is introduced in this article. Multiple secret images can be simultaneously and imperceptibly hid into another cover image with the same image size by the proposed scheme. In order to reduce the volume of secret images to be hid, a codebook is first generated from the secret images and these images are encoded into binary indices by the vector quantization technique. Then the compressed data of secret images are embedded into the VQ codebook used in the encoding procedure by an adaptive least-significant-bits (LSB) modification technique. For the security purpose, the slightly modified codebook is further encrypted into a meaningless data stream by the DES cryptosystem. Finally, the encrypted codebook is embedded into the cover image using the greedy LSB modification technique. Simulation results show that the proposed scheme provides a good improvement in the visual quality of the extracted secret images and the cover image at the receiver. In addition, the proposed scheme provides better hiding capacity for the cover image than earlier works.

Speech Coding and Recognition

This paper investigates the performance of a speech recognizer in an interactive voice response system for various coded speech signals, coded by using a vector quantization technique namely Multi Switched Split Vector Quantization Technique. The process of recognizing the coded output can be used in Voice banking application. The recognition technique used for the recognition of the coded speech signals is the Hidden Markov Model technique. The spectral distortion performance, computational complexity, and memory requirements of Multi Switched Split Vector Quantization Technique and the performance of the speech recognizer at various bit rates have been computed. From results it is found that the speech recognizer is showing better performance at 24 bits/frame and it is found that the percentage of recognition is being varied from 100% to 93.33% for various bit rates. Keywords—Linear predictive coding, Speech Recognition, Voice banking, Multi Switched Split Vector Quantization, Hidden Markov Model, Linear Predictive Coefficients.

System and method for adaptive bit loading source coding via vector quantization

A communication system includes a data transmitter and a data receiver in which the data transmitter transmits data using a multicarrier modulation, such as an OFDM modulation technique. The data receiver determines a predetermined level of modulation for at least one carrier of the multicarrier modulation corresponding to a received signal-to-noise ratio for the carrier using a vector quantization technique. The predetermined level of modulation for each carrier is based on an index to a codeword in a codebook corresponding to the signal-to-noise ratio for the carrier. The index to the codeword is a nearest neighbor to a vector corresponding to the signal-to-noise ratio for the carrier, and the codeword for each predetermined level of modulation is determined based on an adaptive-bit-loading technique. The data receiver transmits the index to the codeword for each carrier to the data transmitter in order to update the modulation at the data transmitter.