An improved vector quantization method using deep neural network

To address the challenging problem of vector quantization (VQ) for high dimensional vector using large coding bits, this work proposes a novel deep neural network (DNN) based VQ method. This method uses a k-means based vector quantizer as an encoder and a DNN as a decoder. The decoder is initialized by the decoder network of deep auto-encoder, fed with the codes provided by the k-means based vector quantizer, and trained to minimize the coding error of VQ system. Experiments on speech spectrogram coding demonstrate that, compared with the k-means based method and a recently introduced DNN-based method, the proposed method significantly reduces the coding error. Furthermore, in the experiments of coding multi-frame speech spectrogram, the proposed method achieves about 11% relative gain over the k-means based method in terms of segmental signal to noise ratio (SegSNR).     

基于高斯-马尔可夫随机场和神经网络的无监督纹理分割

提出一种基于高斯-马尔可夫随机场(GMRF)和神经网络的无监督纹理分割方法,方法分为两步:第一步先将图像分为不重叠的小块,在小块中计算GMRF参数,并将此参数和该块的均值、方差作为该块的特征向量,然后进行聚类,得到原图像的一个初始分割和图像中所包含的类别数;第二步构造一个决定性松弛的神经网络,将第一步得到的结果作为初始输入,经过神经网络计算,得到一个精确的分割结果.实验证明:该方法是一种有效的纹理分割方法.     

电子封装常用名称及术语汇集

电子封装常用名称及术语汇集下面，按英文字母顺序，汇集并解释了与目前ＬＳＩ（包括ＩＣ）正在采用的主要封装形式相关联的名称术语等。这些名称术语参考并引用了日本国内１２个半导体制造公司，其他国家７个半导体制造公司＊与ＬＳＩ封装相关的资料、日本电子机械工业会...     

Randomized neural network based signature for color texture classification

Multidimensional Systems and Signal Processing - Color texture analysis is an important subject in computer vision research. This paper presents an innovative and powerful color texture analysis...     

Intelligent switching control of a pneumatic muscle robot arm using learning vector quantization neural network

Pneumatic cylinders are one of the low-cost actuation sources used in industrial and prosthetic application, since they have a high power/weight ratio, high-tension force and long durability. However, problems with the control, oscillatory motion and compliance of pneumatic systems have prevented their widespread use in advanced robotics. To overcome these shortcomings, a number of newer pneumatic actuators have been developed, such as the McKibben Muscle, Rubber Actuator and Pneumatic Artificial Muscle (PAM) Manipulators. In this paper, the solution for position control of a robot arm with slow motion driven by two pneumatic artificial muscles is presented. However, some limitations still exist, such as a deterioration of the performance of transient response due to the changes in the external load. To overcome this problem, a switching algorithm of the control parameter using a learning vector quantization neural network (LVQNN) is proposed in this paper. The LVQNN estimates the external load of the pneumatic artificial muscle manipulator. The effectiveness of the proposed control algorithm is demonstrated through experiments with different external working loads.     

基于自适应谐振网络的矢量量化研究

简介矢量量化技术,描述了码书设计和码字搜索的原理.分析了自适应谐振网络相对于一般竞争网络的优点.即自适应谐振网络克服了一般竞争网络的稳定性/可塑性困境问题.文章归纳了自适应谐振神经网络的一般结构和学习算法.提出了基于自适应谐振神经网络的码书设计算法,并确定了相关的网络参数.基于自适应谐振神经网络的码书设计比一般竞争网络具有更好的效果.     

Abdominal organ segmentation using texture transforms and aHopfield neural network

Abdominal organ segmentation is highly desirable but difficult, due to large differences between patients and to overlapping grey-scale values of the various tissue types. The first step in automating this process is to cluster together the pixels within each organ or tissue type. We propose to form images based on second-order statistical texture transforms (Haralick transforms) of a CT or MRI scan. The original scan plus the suite of texture transforms are then input into a Hopfield neural network (HNN). The network is constructed to solve an optimization problem, where the best solution is the minima of a Lyapunov energy function. On a sample abdominal CT scan, this process successfully clustered 79-100% of the pixels of seven abdominal organs. It is envisioned that this is the first step to automate segmentation. Active contouring (e.g., SNAKE's) or a back-propagation neural network can then be used to assign names to the clusters and fill in the incorrectly clustered pixels.     

Inverse error-diffusion using classified vector quantization

This correspondence extends and modifies classified vector quantization (CVQ) to solve the problem of inverse halftoning. The proposed process consists of two phases: the encoding phase and decoding phase. The encoding procedure needs a codebook for the encoder which transforms a halftoned image to a set of codeword-indices. The decoding process also requires a different codebook for the decoder which reconstructs a gray-scale image from a set of codeword-indices. Using CVQ, the reconstructed gray-scale image is stored in compressed form and no further compression may be required. This is different from the existing algorithms, which reconstructed a halftoned image in an uncompressed form. The bit rate of encoding a reconstructed image is about 0.51 b/pixel.     

Structured vector quantization using linear transforms

This paper presents novel structured vector quantization (VQ) techniques characterized by the use of linear transformations for the input VQ. The first technique is called the affine transformations VQ, in which the quantized vector is formed by adding the transformed outputs of a multistage codebook rather than just adding the outputs of the stages as in regular multistage vector quantization (MSVQ). The name of the VQ technique comes from the fact that in the two-stage case, the quantized vector is obtained as the result of an affine transformation. This technique can be viewed as a generalized form of MSVQ. If the transformations are constrained to be the identity transformation, this technique becomes identical to the regular MSVQ. The transformations in the introduced technique are selected from a family of linear transformations, represented by a codebook of matrices. In order to I reduce the memory required for storing the matrices, the paper discusses a second technique called scaled rotation matrices VQ, where matrices are constrained to be scaled rotation matrices. Since rotation matrices can be stored by just storing the corresponding rotation angles, this approach enables efficient storage of linear transforms. The design algorithms are based on joint optimization of the linear transformation and the stage codebooks. Experimental results based on speech spectrum quantization show that the proposed VQ techniques outperform the MSVQ of the same bit rate.     

Unsupervised texture segmentation using multichannel decompositionand hidden Markov models

In this paper, we describe an automatic unsupervised texture segmentation scheme using hidden Markov models (HMMs). First, the feature map of the image is formed using Laws' micromasks and directional macromasks. Each pixel in the feature map is represented by a sequence of 4-D feature vectors. The feature sequences belonging to the same texture are modeled as an HMM. Thus, if there are M different textures present in an image, there are M distinct HMMs to be found and trained. Consequently, the unsupervised texture segmentation problem becomes an HMM-based problem, where the appropriate number of HMMs, the associated model parameters, and the discrimination among the HMMs become the foci of our scheme. A two-stage segmentation procedure is used. First, coarse segmentation is used to obtain the approximate number of HMMs and their associated model parameters. Then, fine segmentation is used to accurately estimate the number of HMMs and the model parameters. In these two stages, the critical task of merging the similar HMMs is accomplished by comparing the discrimination information (DI) between the two HMMs against a threshold computed from the distribution of all DI's. A postprocessing stage of multiscale majority filtering is used to further enhance the segmented result. The proposed scheme is highly suitable for pipeline/parallel implementation. Detailed experimental results are reported. These results indicate that the present scheme compares favorably with respect to other successful schemes reported in the literature.     

Unsupervised texture segmentation of images using tuned matchedGabor filters

Recent studies have confirmed that the multichannel Gabor decomposition represents an excellent tool for image segmentation and boundary detection. Unfortunately, this approach when used for unsupervised image analysis tasks imposes excessive storage requirements due to the nonorthogonality of the basis functions and is computationally highly demanding. In this correspondence, we propose a novel method for efficient image analysis that uses tuned matched Gabor filters. The algorithmic determination of the parameters of the Gabor filters is based on the analysis of spectral feature contrasts obtained from iterative computation of pyramidal Gabor transforms with progressive dyadic decrease of elementary cell sizes. The method requires no a priori knowledge of the analyzed image so that the analysis is unsupervised. Computer simulations applied to different classes of textures illustrate the matching property of the tuned Gabor filters derived using our determination algorithm. Also, their capability to extract significant image information and thus enable an easy and efficient low-level image analysis will be demonstrated.     

无监督条件下直扩信号特征码型估计方法

针对于直序列扩频(DSSS)信号盲检测过程中,事先对用户信号特征序列未知这一情况,提出一种应用于非合作条件下对单用户DSSS系统的特征序列的自组织特征映射神经网络(SOFM)估计算法[1,2].理论计算表明,该算法比传统的滑动相关检测法运算量小.仿真结果表明,该方法在非合作、低信噪比条件下实现对DSSS的解扩是可行的.同时,由于该方法是基于神经网络的训练和学习,有很强的适应性,是一种自适应解扩方法.     

Beat-based ECG compression using gain-shape vector quantization

An electrocardiogram (ECG) data compression scheme is presented using the gain-shape vector quantization. The proposed approach utilizes the fact that ECG signals generally show redundancy among adjacent heartbeats and adjacent samples. An ECG signal is QRS detected and segmented according to the detected fiducial points. The segmented heartbeats are vector quantized, and the residual signals are calculated and encoded using the AREA algorithm. The experimental results show that with the proposed method both visual quality and the objective quality are excellent even in low bit rates. An average PRD of 5.97% at 127 b/s is obtained for the entire 48 records in the MIT-BIH database. The proposed method also outperforms others for the same test dataset.     

Subband coding of images using vector quantization

A novel two-dimensional subband coding technique is presented that can be applied to images as well as speech. A frequency-band decomposition of the image is carried out by means of 2D separable quadrature mirror filters, which split the image spectrum into 16 equal-rate subbands. These 16 parallel subband signals are regarded as a 16-dimensional vector source and coded as such using vector quantization. In the asymptotic case of high bit rates, a theoretical analysis yields that a lower bound to the gain is attainable by choosing this approach over scalar quantization of each subband with an optimal bit allocation. It is shown that vector quantization in this scheme has several advantages over coding the subbands separately. Experimental results are given, and it is shown the scheme has a performance that is comparable to that of more complex coding techniques     

Image compression using nonlinear pyramid vector quantization

In this paper, we present a new image compression scheme that exploits the VQ technique in a hierarchical nonlinear pyramid structure. We use multistage median filters (MMF) to build the image pyramids. Image pyramids generated by MMF show a better details preservation than the ones generated by Burt's kernel. It is shown that MMF effectively decorrelates the difference pyramids, resulting in smaller first order entropy. Our simulations on natural images show that NPVQ yields a higher SNR as well as better image quality, in comparison with LPVQ. The NPVQ scheme is also appropriate for progressive image transmission.     

Image coding using entropy-constrained residual vector quantization

An entropy-constrained residual vector quantization design algorithm is used to design codebooks for image coding. Entropy-constrained residual vector quantization has several important advantages. It can outperform entropy-constrained vector quantization in terms of rate-distortion performance, memory, and computation requirements. It can also be used to design vector quantizers with relatively large vector sizes and high output rates. Experimental results indicate that good image reproduction quality can be achieved at relatively low bit rates. For example, a peak signal-to-noise ratio of 30.09 dB is obtained for the 512x512 LENA image at a bit rate of 0.145 b/p.     

Video coding scheme using DCT-pyramid vector quantization

A new and effective video coding scheme for contribution quality is proposed. The CMTT/2, a joint committee of CCIR and CCITT, has proposed a video coding scheme (already approved at European level by ETS) working at 34-45 Mbit/s. Basically this proposal includes a DCT transform for spatial correlation removal and motion compensation for temporal correlation removal. The individual transform coefficients are then scalar quantized with a non uniform bit assignment. Starting from the CMTT/2 proposal, the study presents a new video coding scheme designed using a vector quantizer solution instead of the scalar one. Specifically, the pyramid vector quantization (PVQ) has been chosen as the vector quantization method as it is able to reduce the DCT coefficients Laplacian distribution. Simulation results show that the proposed video coding scheme gives the same contribution quality at 22 Mbit/s as the one obtained with the CMTT/2 proposal at 45 Mbit/s.     

Adaptive vector quantization using a self-development neuralnetwork

A neural network model, called SPAN (space partition network), is presented. This model differs from most of the currently seen neural networks in that it allows a network to adapt its structure by adding neurons, killing neurons, and modifying the structural relationships between neurons in the network. An adaptive vector quantization source-coding system based on SPAN is proposed. The major advantage of using SPAN as the codebook of a vector quantizer is that SPAN can capture the local context of the source signal space and map onto a lattice structure. A fast codebook-searching method utilizing the local context of the lattice is proposed, and a coding scheme, called the path coding method, for eliminating the correlation buried in the source sequence is introduced. The performance of the proposed coder is compared to an LBG (Y. Linde, A. Buzo, and R.M. Gray, 1980) coder on synthesized Gauss-Markov sources. Simulation results show that, without using the path coding method, SPAN yields performance similar to an LBG coder; however, if the path coding method is used, SPAN displays a much better performance than the LBG for highly correlated signal sources     

Unsupervised classification of scattering behavior using radarpolarimetry data

The use of an imaging radar polarimeter data for unsupervised classification of scattering behavior is described by comparing the polarization properties of each pixel in an image to that of simple classes of scattering such as even number of reflections, odd number of reflections, and diffuse scattering. For example, when this algorithm is applied to data acquired over the San Francisco Bay area in California, it classifies scattering by the ocean as being similar to that predicted by the class of odd number of reflections, scattering by the urban area as being similar to that predicted by the class of even number of reflections, and scattering by the Golden Gate Park as being similar to that predicted by the diffuse scattering class. It also classifies the scattering by a lighthouse in the ocean and boats on the ocean surface as being similar to that predicted by the even number of reflections class, making it easy to identify these objects against the background of the surrounding ocean     

Unsupervised classification of cell images using pyramid nodelinking

In this communication we describe a segmentation technique which combines two properties in an iterative and hierarchial matter to correctly segment and classify the given cell images. The technique is applied to digital images taken from microscope slides of cultured rat liver cells, and the goal is to classify these cells into one of three possible classes. The first class cells (I) are morphologically normal and stain the darkest. The second class cells (II) are slightly damaged showing both nuclear and cytoplasmic swelling with resultant lessening of staining affinity. The third class cells (III) are markedly damaged as demonstrated by the presence of cytoplasmic vacuolization, or are completely disintegrated. First class cells are classified by taking advantage of their staining affinity; the original gray level image is segmented into four gray levels. The darkest is then classified as type I. Type III cells are classified by using high business as a characteristic; the standard deviation of the original image is segmented into four business levels. The highest level is classified as type III cell. Assuming only the three cell types are present in any given image, the remaining non-background unclassified pixels are determined to belong to type II cells.