PDVQ图像编码系统的算法研究

PDVQ图像编码系统首先将码书进行方向性分类,把每类方向性码书中的码字按码字和值进行升序排列,并根据EBNNS算法将码书分块。编码时,先根据输入图像块的相关性进行PDVQ编码,然后分析输入图像块的方向性来选择相应的分类子码书,在该子码书中根据输入图像块的和值确定码字搜索范围,最后在确定的搜索范围内搜索最匹配码字。仿真结果表明,该系统集合了动态图像块划分（PDVQ）、基于方向性分类编码和等和值块扩展最近邻码字搜索（EBNNS）三种算法的优点,在保证重建图像质量前提下,缩短了编码时间,并提高了压缩比。     

均值与矢量量化复合快速图像编码算法

矢量量化是一种有效的数据压缩技术,由于其算法简单,具有较高的压缩率,因而被广泛应用于数据压缩编码领域。通过对图像块灰度特征的研究,根据图像的平滑与否,提出了对图像进行均值和矢量量化复合编码算法,该算法对平滑图像块采用均值编码,对非平滑块采用矢量量化编码。这不仅节省了平滑码字的存储空间,提高了码书存储效率,并且编码速度大大提高。同时采用码字旋转反色（2R）压缩算法将码书的存储容量减少到1/8,并结合最近邻块扩展搜索算法（EBNNS）对搜索算法进行优化。在保证图像画质的前提下,整个系统的图像编码速度比全搜索的普通矢量量化平均提高约7.7倍。     

一种新颖的基于哈德码变换的码字搜索算法

针对图像矢量量化编码的复杂性,提出了一种新颖的快速最近邻码字搜索算法。该算法首先计算出每个码字和输入矢量的哈德码变换,然后为输入矢量选取范数距离最近的初始匹配码字,利用多控制点的三角不等式和两条有效的码字排除准则,把不匹配的码字排除,最后选取与输入矢量最匹配的码字。实验结果表明,新算法相比于其他算法,在保证编码质量的前提下,码字搜索时间和计算量均有了明显降低。     

基于矢量量化码书的离群点检测方法*

利用矢量量化码书作为数据分类模式最优代表集的特点,提出基于码书的离群点概念,论证了其与经典统计学关于离群点定义的内在联系。在基于学习的矢量量化码书生成算法和最近邻码字搜索算法基础上构造了离群点检测算法。实验结果表明了提出的关于离群点定义的合理性和算法的有效性。     

初始码字间距最大化的矢量量化码书设计算法

目前对传统LBG算法的改进措施一般以增加时间开销作为代价.本文提出一种新的矢量量化码书设计改进措施--初始码字间距最大化:初始码书中的码字全部来自输入的训练矢量,且每一个新的初始码字尽可能地远离现有的码字,实验结果表明:本算法完全消除了空胞腔现象,更有效地避免了局部最优,能获得质量更高的码书;收敛速度快,具有较低的时间消耗.本算法在时间开销以及码书质量这两个方面都优于传统LBG和基于人工蚁群优化的码书设计算法等改进算法.     

基于改进人工蜂群的图像矢量量化码书设计算法

为了进一步提高图像矢量量化的码书质量,提出了一种新的图像压缩矢量量化码书设计算法。该算法采用均方误差(MSE)作为码书设计的适应度函数,利用改进的人工蜂群算法进行适应度函数的优化求解,增强了算法的自组织性和收敛性,大大减少了陷入局部收敛的可能性。将一种基于和值特性的快速码字搜索思想引入到码书设计算法中,使算法计算量明显降低。仿真结果表明,该算法具有计算时间短、收敛速度快的优点,并且生成的码书质量好、稳定性强。     

码书排序对快速码字搜索算法性能影响的分析

矢量量化快速码字搜索算法中,为了有效地减小搜索范围,必须对原始码书按一定的准则进行重新排序。对现存的两类快速码字搜索算法进行了总结,其中一类是码书按1维顺序关系排序,另一类是码书按2维相邻关系排序。通过实验给出了两类算法的搜索范围和编码时间,并进行了比较和分析,进而提出了在实际编码时如何更好地使用这两种排序关系的准则。     

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。     

利用PNN算法改进初始码书的GLA算法

针对改进广义Lloyd算法（GLA）对初始码书的敏感性，提出用成对最近邻（PNN）算法训练码书作为GLA算法的初始码书，再由GLA算法产生最终码书。PNN算法得到的码书作为GLA算法的初始码书，其码字能在整个输入矢量空间中很好地散开，尽可能地占据输入概率密度较大区域，提高最后码书的质量。仿真实验中，利用正态分布的随机数训练码书，恢复一段正态分布的随机数，采用Mahalanobis失真测度评价恢复数据的失真度。仿真结果表明，改进算法降低了GLA算法对初始码书的敏感性，提高了最终训练码书的质量，降低了恢复数据的失真。     

基于新的空胞腔策略的矢量量化码书设计方法

为了克服传统LBG算法中的空胞腔现象,提出了一种基于码字间距最大化的新的空胞腔策略。利用离码书距离最大的输入矢量来修改胞腔中的码字,旨在形成码字的合理分布,减小矢量量化的平均失真。实验结果表明：提出的策略能有效地消除空胞腔现象,获得性能较好的码书,其峰值信噪比比传统的LBG算法提高了3 dB。     

Hadamard transform based fast codeword search algorithm for high-dimensional VQ encoding

An efficient nearest neighbor codeword search algorithm for vector quantization based on the Hadamard transform is presented in this paper. Four elimination criteria are derived from two important inequalities based on three characteristic values in the Hadamard transform domain. Before the encoding process, the Hadamard transform is performed on all the codewords in the codebook and then the transformed codewords are sorted in the ascending order of their first elements. During the encoding process, firstly the Hadamard transform is applied to the input vector and its characteristic values are calculated; secondly, the codeword search is initialized with the codeword whose Hadamard-transformed first element is nearest to that of the input vector; and finally the closest codeword is found by an up-and-down search procedure using the four elimination criteria. Experimental results demonstrate that the proposed algorithm is much more efficient than the most existing nearest neighbor codeword search algorithms in the case of problems of high dimensionality.     

使用2－范数匹配的快速分形图像编码算法

尽管分形图像压缩在高压缩比时具有高的重建图像质量,但由于它编码过程耗时而未能在图像压缩领域广泛应用。为了提高分形图像编码过程的速度,提出了基于2－范数匹配的快速分形图像编码算法,该算法先把码书里的码块按2－范数大小赋序,然后对每个输入Range块,在赋序码书中寻找与它的2－范数最接近的码块,进而在这个码块的领域内搜索Range块的最佳匹配块。计算机仿真结果显示,在不影响基本分形图像编码算法解码图像质量的情况下,该方案大大加快了它的编码速度。     

Accelerating VQ-based codeword search on the basis of partial search strategy

In this paper, a technique for accelerating the search on VQ-based codeword search is proposed. With our approach, all the pixel blocks of vector representation in an image picture could be encoded efficiently into their corresponding indices, and be associated with the closest codeword in the pre-generated codebook. The technique adopted in our scheme is inspired by the concept of space partition of the initial codebook. It is accomplished in a manner that the search range for the image block is significantly reduced. There is a key-codebook comprised of numerous key-codewords, and with a smaller book size, which is generated from the given codewords during system initialization. Any image block is then directed to look for the closest key-codeword in the key-codebook. Ultimately, the best-match codeword is checked out according to the relation between the closest key-codeword and the ‘genuine’ codewords in the given codebook. This short-time achievement is obtained because of the considerable book size reduction. A flexible radius, spread by a key-codeword is imposed in our elaborated algorithm to attain the most precise hit ratio estimation. The experiments show that our scheme is at least two and a half times faster than that of a full search in VQ implementation. Moreover, the strategy we proposed is also compatible with the search algorithms in finding the closest codeword, and the high quality of image display remained the same.     

An improved search algorithm for vector quantization using mean pyramid structure

Vector quantization (VQ) is a well-known data compression technique. In the codebook design phase as well as the encoding phase, given a block represented as a vector, searching the closest codeword in the codebook is a time-consuming task. Based on the mean pyramid structure and the range search approach, an improved search algorithm for VQ is presented in this paper. Conceptually, the proposed algorithm has the bandpass filter effect. Each time, using the derived formula, the search range becomes narrower due to the elimination of some portion of the previous search range. This reduces search times and improves the previous result by Lee and Chen (A fast search algorithm for vector quantization using mean pyramids of codewords. IEEE Trans. Commun. 43(2/3/4), (1995) 1697–1702). Some experimental results demonstrate the computational advantage of the proposed algorithm.     

Fast additive quantization for vector compression in nearest neighbor search

Vector quantization has been widely employed in nearest neighbor search because it can approximate the Euclidean distance of two vectors with the table look-up way that can be precomputed. Additive quantization (AQ) algorithm validated that low approximation error can be achieved by representing each input vector with a sum of dependent codewords, each of which is from its own codebook. However, the AQ algorithm relies on computational expensive beam search algorithm to encode each vector, which is prohibitive for the efficiency of the approximate nearest neighbor search. In this paper, we propose a fast AQ algorithm that significantly accelerates the encoding phase. We formulate the beam search algorithm as an optimization of codebook selection orders. According to the optimal order, we learn the codebooks with hierarchical construction, in which the search width can be set very small. Specifically, the codewords are firstly exchanged into proper codebooks by the indexed frequency in each step. Then the codebooks are updated successively to adapt the quantization residual of previous quantization level. In coding phase, the vectors are compressed with learned codebooks via the best order, where the search range is considerably reduced. The proposed method achieves almost the same performance as AQ, while the speed for the vector encoding phase can be accelerated dozens of times. The experiments are implemented on two benchmark datasets and the results verify our conclusion.     

基于预测码矢的快速搜索新算法

在ｋ维欧氏空间Ｒｋ中，给定一个有限子集Ｗ及一个向量ｘ，如何搜索Ｗ中与ｘ距离最近的向量，具有重要的实际应用价值，尤其在图象的矢量量化编码、神经网络模式识别［１］等问题中，快速搜索起决定性的作用。在分析已有快速搜索算法的基础上，给出一种新的快速搜索算法，该算法利用图象相邻块的码书地址，作为当前块的预测值，使搜索空间缩小更快。     

主对角和特征的快速分形图像编码

分形图像编码十余年来在图像处理尤其是图像压缩领域引起了人们的极大兴趣。但编码过程耗时长限制了它的应用范围。为了缩短编码时间,根据匹配均方根误差与新定义的规范块主对角和特征间的关系,提出了一个限制搜索空间的算法：对一个待编码range块,仅在与该range块主对角和特征值最接近的domain块的邻域范围内搜索它的最佳匹配块。同时融入两个措施：一是预先从码书Ω中排除小标准差domain块;二是对小方差range块用其均值块代替。该算法不仅从理论上证明是可行的,而且三幅标准测试图像的仿真实验结果也表明,它确实能够在重建图像质量略好的情况下,平均加快全搜索分形图像编码算法的编码速度30余倍。     

应用多索引加法量化编码的近邻检索算法

目的 海量图像检索技术是计算机视觉领域研究热点之一,一个基本的思路是对数据库中所有图像提取特征,然后定义特征相似性度量,进行近邻检索。海量图像检索技术,关键的是设计满足存储需求和效率的近邻检索算法。为了提高图像视觉特征的近似表示精度和降低图像视觉特征的存储空间需求,提出了一种多索引加法量化方法。方法 由于线性搜索算法复杂度高,而且为了满足检索的实时性,需把图像描述符存储在内存中,不能满足大规模检索系统的需求。基于非线性检索的优越性,本文对非穷尽搜索的多索引结构和量化编码进行了探索新研究。利用多索引结构将原始数据空间划分成多个子空间,把每个子空间数据项分配到不同的倒排列表中,然后使用压缩编码的加法量化方法编码倒排列表中的残差数据项,进一步减少对原始空间的量化损失。在近邻检索时采用非穷尽搜索的策略,只在少数倒排列表中检索近邻项,可以大大减少检索时间成本,而且检索过程中不用存储原始数据,只需存储数据集中每个数据项在加法量化码书中的码字索引,大大减少内存消耗。结果 为了验证算法的有效性,在3个数据集SIFT、GIST、MNIST上进行测试,召回率相比近几年算法提升4%~15%,平均查准率提高12%左右,检索时间与最快的算法持平。结论 本文提出的多索引加法量化编码算法,有效改善了图像视觉特征的近似表示精度和存储空间需求,并提升了在大规模数据集的检索准确率和召回率。本文算法主要针对特征进行近邻检索,适用于海量图像以及其他多媒体数据的近邻检索。