Divergence-based classification in learning vector quantization

We discuss the use of divergences in dissimilarity-based classification. Divergences can be employed whenever vectorial data consists of non-negative, potentially normalized features. This is, for instance, the case in spectral data or histograms. In particular, we introduce and study divergence based learning vector quantization (DLVQ). We derive cost function based DLVQ schemes for the family of γ‐divergences which includes the well-known Kullback-Leibler divergence and the so-called Cauchy-Schwarz divergence as special cases. The corresponding training schemes are applied to two different real world data sets. The first one, a benchmark data set (Wisconsin Breast Cancer) is available in the public domain. In the second problem, color histograms of leaf images are used to detect the presence of cassava mosaic disease in cassava plants. We compare the use of standard Euclidean distances with DLVQ for different parameter settings. We show that DLVQ can yield superior classification accuracies and Receiver Operating Characteristics.     

Efficient music note recognition based on a self-organizing map tree and linear vector quantization

Using classical signal processing and filtering techniques for music note recognition faces various kinds of difficulties. This paper proposes a new scheme based on neural networks for music note recognition. The proposed scheme uses three types of neural networks: time delay neural networks, self-organizing maps, and linear vector quantization. Experimental results demonstrate that the proposed scheme achieves 100% recognition rate in moderate noise environments. The basic design of two potential applications of the proposed scheme is briefly demonstrated.     

Classified self-organizing map with adaptive subcodebook for edge preserving vector quantization

This paper presents a novel classified self-organizing map method for edge preserving quantization of images using an adaptive subcodebook and weighted learning rate. The subcodebook sizes of two classes are automatically adjusted in training iterations based on modified partial distortions that can be estimated incrementally. The proposed weighted learning rate updates the neuron efficiently no matter of how large the weighting factor is. Experimental results show that the new method achieves better quality of reconstructed edge blocks and more spread out codebook and incurs a significantly less computational cost as compared to the competing methods.     

Fuzzy learning vector quantization for hyperspectral coastal vegetation classification

Artificial neural networks (ANNs) may be of significant value in extracting vegetation type information in complex vegetation mapping problems, particularly in coastal wetland environments. Unsupervised, self-organizing ANNs have not been employed as frequently as supervised ANNs for vegetation mapping tasks, and further remote sensing research involving fuzzy ANNs is also needed. In this research, the utility of a fuzzy unsupervised ANN, specifically a fuzzy learning vector quantization (FLVQ) ANN, was investigated in the context of hyperspectral AVIRIS image classification. One key feature of the neural approach is that unlike conventional hyperspectral data processing methods, endmembers for a given scene, which can be difficult to determine with confidence, are not required for neural analysis. The classification accuracy of FLVQ was comparable to a conventional supervised multi-layer perceptron, trained with backpropagation (MLP) (KHAT (Kˆ) accuracy: 82.82% and 84.66%, respectively; normalized accuracy: 74.60% and 75.85%, respectively), with no significant difference at the 95% confidence level. All neural algorithms in the experiment yielded significantly higher classification accuracies than the conventional endmember-based hyperspectral mapping method assessed (i.e., matched filtering, where Kˆ accuracy = 61.00% and normalized accuracy = 57.96%). FLVQ was also dramatically more computationally efficient than the baseline supervised and unsupervised ANN algorithms tested, including the MLP and the Kohonen self-organizing map (SOM), respectively. The 400-neuron FLVQ network required only 3.6% of the computation time used by the MLP network, and only 5.9% of the MLP time was used by the 588-neuron FLVQ network. In addition, the 400-neuron FLVQ used only 16.7% of the time used by the 400-neuron SOM for model development.     

基于矢量量化和深度优先搜索的轨迹分布模式学习算法

在对轨迹流矢量进行量化编码的基础上,提出了一种基于深度优先搜索的轨迹分布模式提取算法,生成了能够描述轨迹分布的序列模式图,并给出了与之相应的异常检测和行为预测方法。对不同场景的可见光和红外序列图像的实验表明,本文方法不仅能够学习轨迹中流矢量的分布,而且能够反映它们之间的时序关系,可以应用于室外复杂场景的目标异常行为检测。     

Improved transmission of vector quantized data over noisy channels

The conventional channel-optimized vector quantization (COVQ) is very powerful in the protection of vector quantization (VQ) data over noisy channels. However, it suffers from the time consuming training process. A soft decoding self-organizing map (SOM) approach for VQ over noisy channels is presented. Compared with the COVQ approach, it does not require a long training time. For AWGN and fading channels, the distortion of the proposed approach is comparable to that of COVQ. Simulation confirmed that our proposed approach is a fast and practical method for VQ over noisy channels.     

Improved batch fuzzy learning vector quantization for image compression

In this paper, we develop a batch fuzzy learning vector quantization algorithm that attempts to solve certain problems related to the implementation of fuzzy clustering in image compression. The algorithm’s structure encompasses two basic components. First, a modified objective function of the fuzzy c-means method is reformulated and then is minimized by means of an iterative gradient-descent procedure. Second, the overall training procedure is equipped with a systematic strategy for the transition from fuzzy mode, where each training vector is assigned to more than one codebook vectors, to crisp mode, where each training vector is assigned to only one codebook vector. The algorithm is fast and easy to implement. Finally, the simulation results show that the method is efficient and appears to be insensitive to the selection of the fuzziness parameter.     

Co-adaptation of self-organizing maps by evolution and learning

This paper proposes two co-adaptation schemes of self-organizing maps that incorporate the Kohonen's learning into the GA evolution in an attempt to find an optimal vector quantization codebook of images. The Kohonen's learning rule used for vector quantization of images is sensitive to the choice of its initial parameters and the resultant codebook does not guarantee a minimum distortion. To tackle these problems, we co-adapt the codebooks by evolution and learning in a way that the evolution performs the global search and makes inter-codebook adjustments by altering the codebook structures while the learning performs the local search and makes intra-codebook adjustments by making each codebook's distortion small. Two kinds of co-adaptation schemes such as Lamarckian and Baldwin co-adaptation are considered in our work. Simulation results show that the evolution guided by a local learning provides the fast convergence, the co-adapted codebook produces better reconstruction image quality than the non-learned equivalent, and Lamarckian co-adaptation turns out more appropriate for the VQ problem.     

Data preprocessing for adaptive vector quantization

This paper describes a method for designing a codebook for vector quantization (VQ), based on preprocessing of the input data which makes them block-stationary, and on a criterion which takes into account the error visibility of the image to be coded. Test results, carried out at about 1.2 bits/pel bit rate, indicate that the proposed VQ enables reconstruction of images (both outside and inside the training set) with very low distortion, and exhibits high robustness, the variance of the SNR being sensibly lower than in the case of unprocessed data.     

基于向量量化的图像超分辨率方法

受分形编码思想启发,提出了一种新的基于向量量化的图像超分辨率方法。该方法使用学习算法来获取单幅输入图像中的高频信息和低频信息之间的对应关系,并利用此关系对输入图像的一个倍频程的空间频率内添加图像细节以获得高分辨率图像。该方法克服了传统插值方法中因过度平滑导致图像模糊和纹理保持较差的缺点,能够重现出传统插值方法不能复原出的一些高频图像细节。实验结果显示该算法在客观和主观上都比传统插值方法有更好的评价。     

矢量量化压缩算法在高光谱影像上的研究实现

针对高光谱影像光谱维的数据量大、传统影像压缩方法不易于保存光谱内信息的特点,对矢量量化数据压缩方法中码书设计和码字搜索两个关键技术进行详细地研究,提出针对高光谱影像压缩的改进方法,并在此基础上实现了对高光谱影像的矢量量化压缩算法。最后通过对不同波段组合的AVIRIS的高光谱数据的实验,从压缩后的压缩率、速率和失真率等方面进行观察和对比,证明矢量量化压缩算法对高光谱影像具有显著的压缩效果。     

Drill flank wear estimation using supervised vector quantization neural networks

Drill wear detection and prognosis is one of the most important considerations in reducing the cost of rework and scrap and to optimize tool utilization in hole making industry. This study presents the development and implementation of two supervised vector quantization neural networks for estimating the flank-land wear size of a twist drill. The two algorithms are; the learning vector quantization (LVQ) and the fuzzy learning vector quantization (FLVQ). The input features to the neural networks were extracted from the vibration signals using power spectral analysis and continuous wavelet transform techniques. Training and testing were performed under a variety of speeds and feeds in the dry drilling of steel plates. It was found that the FLVQ is more efficient in assessing the flank wear size than the LVQ. The experimental procedure for acquiring vibration data and extracting features in the time-frequency domain using the wavelet transform is detailed. Experimental results demonstrated that the proposed neural network algorithms were effective in estimating the size of the drill flank wear.     

Implementation of a vector quantization codebook design technique based on a competitive learning artificial neural network

We describe an implementation of a vector quantization codebook design algorithm based on the frequencysensitive competitive learning artificial neural network. The implementation, designed for use on high-performance computers, employs both multitasking and vectorization techniques. A C version of the algorithm tested on a CRAY Y-MP8/864 is discussed. We show how the implementation can be used to perform vector quantization, and demonstrate its use in compressing digital video image data. Two images are used, with various size codebooks, to test the performance of the implementation. The results show that the supercomputer techniques employed have significantly decreased the total execution time without affecting vector quantization performance.This work was supported by a Cray University Research Award and by NASA Lewis research grant number NAG3-1164.     

直接矢量量化方法在语音编码算法中的应用

介绍了一种降低码书搜索复杂度的方法-直接矢量量化（DVQ）方法,将其应用于LD-CELP语音编码算法中的仿真译码器模块和码书搜索模块,用感觉加权逆滤波器代替仿真译码器模块中的综合滤波器,去除了码书搜索模块中冲激响应h（n）的运算。实验结果表明,利用直接矢量量化方法简化了码书搜索算法的复杂度,提高了码书搜索算法的效率,在运算时间方面比原始LD-CELP算法快3 s~5 s,同时保持了原编码算法合成语音的音质。     

一种基于索引约束矢量量化的脆弱音频水印算法*

与传统矢量量化不同,索引约束矢量量化在量化过程中通过约束码字索引二进制形式中某一位的值来限定码字的搜索范围。本文利用其特殊的码字搜索方法提出了一种在音频信号中嵌入水印的方法。将原始音频信号分段,每段进行DCT变换并提取若干中频系数构成矢量。水印嵌入时根据水印比特信息和预先设定的索引约束位的值找到匹配码字修改各段DCT中频系数。水印提取时利用传统矢量量化方法得到各量化索引值后,提取出各索引值中与嵌入端相同位的比特值即为水印信息。该方法在量化过程中嵌入水印信息,有很好的实时性。实验结果表明,利用该方法嵌入的水印为一种脆弱水印,可用于认证。     

有限状态矢量量化在语音端点检测中的应用

语音端点检测在语音处理中占有非常重要的地位,传统的检测方法是基于短时能量和过量率的双门限比较法,但是在信噪比较低的情况下,利用短时能量和过量率很难得到准确的检测结果。另外,在双门限比较法中,判别门限的取值对整个端点的检测影响很大,而这个门限值往往是靠经验所得,具有不稳定性。因此,针对传统方法的不足,根据语音帧间相关性,提出了一种改进算法。让语音信号通过双门限比较,完成端点检测的一级粗判,在语音起止点的模糊帧段,取一定范围的信号矢量,让这些矢量经过处理后再通过有限状态矢量量化器(FSVQ),得到量化矢量,再对量化矢量进行二级细判,从而得到准确的语音起止点。将改进算法应用于汉语连续数字语音识别,平均识别时间由原来的0.871s缩短为0.719s,平均识别率由原来的81.47%上升至89.13%,实验结果表明了该算法的有效性。     

矢量量化中码书旋转压缩的研究

普通码书中的码字之间在不同的方向上具有很大的相关性,存在大量的数据冗余。提出了将码书中的码字旋转压缩的理论。该理论是将各个码字按四个方向垂直旋转后进行相似性检查。如果旋转后的码字其中一个方向上与前面的码字存在相似,则将该码字删除,从而达到压缩的目的。编码时将压缩后的码书旋转恢复后进行编码,从而大幅降低了需要存储的码字数量。同时给出了一种将现有1 024阶16维码书旋转压缩成256阶16维的方法,并对该方法得到的码书性能进行了仿真验证。实验结果表明使用压缩后的码书在硬件实现时与普通的矢量量化码书相比减少了75%的存储空间和输入带宽,而PSNR平均只降低0.28 dB。     

自适应量化权重用于通信高效联邦学习

针对联邦学习训练过程中通信资源有限的问题,本文提出了两种联邦学习算法:自适应量化权重算法和权重复用控制算法,前者对权重的位数进行压缩,减少通信过程中传输的比特数,算法在迭代过程中,自适应调整量化因子,不断减少量化误差;后者能阻止不必要的更新上传,从而减少上传的比特数.基于标准检测数据集Mnist和Cifar10,在CNN和MLP网络模型上做了仿真模拟,实验结果表明,与典型的联邦平均算法相比,提出的算法降低了75%以上的通信成本.     

基于矢量量化的弦乐颤音自动检测

自动音乐标注是音频信息检索的基础,并可广泛应用于辅助音乐教学、辅助音乐创作等许多音乐相关领域。然而,在弦乐器演奏的音乐中存在着大量用于修饰或表现乐曲情感、风格的颤音。在对这类乐器的自动音乐标注中,如果不进行颤音检测而直接进行标注容易出现错误。对基于矢量量化的弦乐颤音识别方法进行了研究,提出了一种用于对整段音乐进行颤音检测的方法。实验证明这种方法是精确有效的。     

应用协同进化的图像矢量量化码书设计方法

为了减小LBG算法对初始码书的依赖性,提高跳出局部最优的能力,提出了一种基于协同进化的矢量量化码书设计方法（Coevolution Based LBG,CLBG）。该算法根据码书在同其他码书竞争中的表现来衡量码书的适应度。实验结果表明：CLBG有效地减小了算法对初始码书的依赖性,所得码书性能超过了其他典型的改进码书设计方法。