应用ICA滤波器技术提取图像纹理特征

针对纹理图像分类问题,本文提出了一种应用ICA滤波器技术提取图像纹理特征的方法.该方法首先从训练图像集中随机抽取图像块作为观测信号,应用ICA技术,提取滤波器组.然后根据训练样本图像对滤波器组的响应值来评估和选择滤波器组,达到降维的目的.最后利用滤波器组对测试图像进行滤波,得到该图像的滤波响应结果,从该响应结果中得到最大响应滤波器编号,提取其直方图作为图像的全局特征和局部特征.对Brodatz纹理图像集中108个纹理类别进行了分类实验,结果表明,与MPEG-7纹理描述子相比,该图像特征对纹理图像具有更好的分类效果.     

基于DT-CWT和SVM的纹理分类算法

提出了一种基于双树复数小波变换(DT-CWT)和支持向量机(SVM)的纹理分类算法.双树复数小波变换不仅具有实数小波的诸多优点,而且还具有近似平移不变性、良好的方向选择性和低冗余度,并且能对图像进行完全重构,能够更好地刻画纹理的特性;支持向量机算法是近年发展起来的性能优越的分类算法,比传统分类器有很大的优越性:避免了局部最优解和"维数灾"问题,其最优分类超平面的思想能够提高分类准确度.该方法用双树复数小波对纹理图像进行滤波并在各方向子带上进行重构,再计算其局部能量函数得到每个像素的特征向量,最后利用支持向量机算法实现对纹理图像像素的分类.将本方法与其它的分类算法进行比较,实验结果表明,提出的算法能有效地提高正确分类率.     

基于Gabor滤波系数高阶矩的图像检索

在分析Gabor滤波器进行图像纹理特征提取的基础上,提出了利用多尺度和多方向Gabor滤波系数的高阶矩提取图像特征进行CBIR的方法,利用滤波系数的方差给出了基于Gabor滤波组提取的图像纹理特征的平滑度和纹理一致性算法,并采用四个尺度和六个方向的滤波系数的能量、方差、峰态、平滑度和一致性组成了CBIR特征向量.采用Brodatz纹理库和Corel图像库中的典型图像进行了对比实验.实验证明,提出的方法比传统的Gabor滤波进行CBIR具有更高的查准率.     

图像调焦算法研究及FPGA实现

利用图像处理方法进行自动调焦的关键是提取图像清晰度特征,并建立其评价算法.本文研究了灰度值线性变换、灰度直方图均衡、中值滤波及同态滤波等预处理方法和基于功率谱的清晰度评价函数,并与其它的评价方法进行了比较分析.研究表明,中值滤波和灰度值线性变换相结合的预处理方法,具有效果好、计算量少等优点;基于功率谱的清晰度评价函数比其它的评价方法具有更好的调焦性能和更明确的物理意义.根据基于功率谱的图像调焦算法的特点,设计了图像处理模块的结构框架和算法流程,提出了流水线作业结构、"乒乓"操作模式、双蝶形处理器复用、基-2 FFT算法的FPGA实现方案,提高了图像自动调焦的计算和响应速度.     

基于频谱滤波的多纹理提取算法

研究了一种在有多种纹理叠加的复杂图像中进行单一纹理提取的算法.首先采用极坐标方法分析纹理频谱的环型和楔型特征并求出纹理分布的周期和方向特征;然后根据这些特征在频域中构建环型和楔型Gauss带通滤波器对纹理频谱进行滤波;再将滤波后的频谱图像转换到时域中就得到了只保留相应纹理成分的图像;最后经过纹理区域矫正处理后就可以提取出真实的纹理.实验结果表明,该方法可以准确提取出图像中叠加的多种纹理,并能完整保留每种纹理的基本特征,在纹理分布不均匀的区域也能提取出纹理的骨架.     

基于系数关系共生矩阵和SVM的纹理分析

纹理是图像中非常重要的特征.提出了一种新的纹理特征提取算法,即对纹理图像进行离散小渡框架变换后,利用同一变换尺度下的小波高频系数与低频系数之间的依存关系信息,构造系数共生矩阵,在此基础上进行纹理特征提取,而不是独立地提取各子带系数特征.考虑支撑向量机(SVM)在小样本数据库和泛化能力方面的优势,在分类实验中采用支撑向量机分类器,实验结果表明,基于这种共生矩阵特征提取分类算法能得到很好的分类结果.     

自适应红外图像直方图均衡增强算法

针对红外图像的特点,本文提出了一种自适应红外图像直方图均衡增强算法.该方法根据原始图像的直方图,自适应地构造出一个加权函数对原始图像的直方图进行加权处理,然后采用加权后的直方图对原始图像进行直方图均衡化.算法不需要人为指定阈值,而且克服了传统直方图均衡提升红外图像背景的缺点.实验结果表明,该算法对红外图像具有较好的增强效果,能够有效地抑制图像的背景,突出目标.     

基于核竞争学习算法的图像特征提取

利用核函数学习可有效解决图像特征线性不可分的特性,结合稀疏表示算法的优势,提出了一种新的图像特征提取方法。采用基于竞争学习规则的独立分量分析法对图像进行稀疏表示,该算法可提取数据的高维特征,且不需要优化高阶的非线性函数和进行稀疏密度估计,因而有较快的收敛速度。与仅使用基于竞争学习的独立分量分析法相比,在PolyU数据库上的实验结果表明,采用基于核函数学习和稀疏表示相结合的方法所提取的数据特征有利于提高特征分类精度。     

用Gabor滤波实现虹膜图像纹理编码

针对虹膜纹理的幅度和相位信息特征,利用二维Gabor滤波器族可以很好地模拟皮层简单细胞的二维感受野轮廓,能够最大限度的提供图像局部方向和频率信息的特点,提出一种基于二维Gabor函数族对虹膜图像进行子块分解滤波的改进编码算法。实验结果表明,该方法能有效的提取虹膜纹理特征,从而达到虹膜识别的目的。     

基于区域灰度变化的虹膜识别

为了提高虹膜识别的效率和准确性,对虹膜识别进行了深入研究,提出了一种基于区域灰度变化的虹膜识别方法.图像预处理得到归一化的虹膜图像,根据虹膜纹理尺度的不同,采用椭圆形高斯滤波器对虹膜图像进行滤波,均匀选取采样点,计算每个采样点附近的区域灰度值,利用序数测度的方法对特征编码,最后通过两个虹膜编码之间的海明距来判断两者之间的差异程度,得到识别结果.本文算法选用Bath 大学虹膜库和CASIA 虹膜库进行了验证实验,等错率分别达到0.05%和0.69%.结果表明,该算法具有很好的准确性和识别速度.     

独立分量分析的图像融合算法

独立分量分析可实现图像的稀疏编码并具有能很好地捕捉图像重要边缘信息的特性.本文提出一种基于独立分量分析的图像融合算法,结合支持向量机对多聚焦图像的清晰域、模糊域进行判断以及在ICA域中进行图像分割以提取图像的主要边缘特征信息来实现特征级的多聚焦图像的融合.实验结果表明,本文提出的融合算法是有效的.     

基于光谱特征和纹理特征协同学习的高光谱图像数据分类

高光谱遥感图像中包含有大量的高维数据,传统的有监督学习算法在对这些数据进行分类时要求获取足够多的有标记样本用于分类器的训练.然而,对高光谱图像中大量的复杂地物像元所属类别进行准确标注通常需要耗费极大的人力.在本文中,我们提出了一种基于半监督学习的光谱和纹理特征协同学习(STF-CT)--法,利用协同学习机制将高光谱图像光谱特征和空间纹理特征这两种不同的特征结合起来,用于小训练样本集下的高光谱图像数据分类问题.STF-CT算法充分利用了高光谱图像的光谱和纹理特征这两个独立视图,构建起一种有效的半监督分类方法,用于提升分类器在小训练样本集情况下的分类精度.实验结果表明该算法在小训练样本集下的高光谱地物分类问题上具有很好的效果.     

基于纹理特征的钢丝绳图像分割方法

针对复杂背景下钢丝绳图像难以准确分割的问题,提出一种新的基于纹理特征的图像分割方法.首先,采用局部二进制模式(Local Binary Pattern,LBP)特征直方图的一阶熵、二阶熵作为LBP特征的统计测度,降低LBP特征的维数.同时选用边缘密度作为纹理描述的特征之一,弥补LBP算子提取纹理特征不足,抗干扰能力差的缺点.然后以上述纹理特征构成特征矢量,采用模糊C-均值(Fuzzy C-Mean,FCM聚类算法进行聚类分割.在实验中,对比了该算法与灰度共生矩阵、传统LBP算子在钢丝绳图像分割中的效果.结果表明,该算法可以有效地对钢丝绳图像进行纹理分割,并能取得良好的边界定位效果,性能优于另外两种算法.     

Application of core vector machines for on-line voltage security assessment using a decisiontree-based feature selection algorithm

This study presents a core vector machine (CVM)-based algorithm for on-line voltage security assessment of power systems. To classify the system security status, a CVM has been trained for each contingency. The proposed CVM-based security assessment algorithm has a very small training time and space in comparison with support vector machines (SVMs) and artificial neural networks (ANNs)-based algorithms. The proposed algorithm produces less support vectors (SVs). Therefore is faster than existing algorithms. One of the main points to apply a machine learning method is feature selection. In this study, a new decision tree (DT)-based feature selection algorithm has been presented. The proposed CVM algorithm has been applied to New England 39-bus power system. The simulation results show the effectiveness and the stability of the proposed method for on-line voltage security assessment. The effectiveness of the proposed feature selection algorithm has also been investigated. The proposed feature selection algorithm has been compared with different feature selection algorithms. The simulation results demonstrate the effectiveness of the proposed feature algorithm.     

A new approach to simulate characterization of particulate matter employing support vector machines

This paper, for the first time, applies the support vector machines (SVMs) paradigm to identify the optimal segmentation algorithm for physical characterization of particulate matter. Size of the particles is an essential component of physical characterization as larger particles get filtered through nose and throat while smaller particles have detrimental effect on human health. Typical particulate characterization processes involve image reading, preprocessing, segmentation, feature extraction, and representation. Of these various steps, knowledge based selection of optimal image segmentation algorithm (from existing segmentation algorithms) is the key for accurately analyzing the captured images of fine particulate matter. Motivated by the emerging machine-learning concepts, we present a new framework for automating the selection of optimal image segmentation algorithm employing SVMs trained and validated with image feature data. Results show that the SVM method accurately predicts the best segmentation algorithm. As well, an image processing algorithm based on Sobel edge detection is developed and illustrated.     

Comparative evaluation of despeckle filtering in ultrasound imaging of the carotid artery

It is well-known that speckle is a multiplicative noise that degrades the visual evaluation in ultrasound imaging. The recent advancements in ultrasound instrumentation and portable ultrasound devices necessitate the need of more robust despeckling techniques for enhanced ultrasound medical imaging for both routine clinical practice and teleconsultation. The objective of this work was to carry out a comparative evaluation of despeckle filtering based on texture analysis, image quality evaluation metrics, and visual evaluation by medical experts in the assessment of 440 (220 asymptomatic and 220 symptomatic) ultrasound images of the carotid artery bifurcation. In this paper a total of 10 despeckle filters were evaluated based on local statistics, median filtering, pixel homogeneity, geometric filtering, homomorphic filtering, anisotropic diffusion, nonlinear coherence diffusion, and wavelet filtering. The results of this study suggest that the first order statistics filter lsmv, gave the best performance, followed by the geometric filter gf4d, and the homogeneous mask area filter lsminsc. These filters improved the class separation between the asymptomatic and the symptomatic classes based on the statistics of the extracted texture features, gave only a marginal improvement in the classification success rate, and improved the visual assessment carried out by the two experts. More specifically, filters lsmv or gf4d can be used for despeckling asymptomatic images in which the expert is interested mainly in the plaque composition and texture analysis; and filters lsmv, gf4d, or lsminsc can be used for the despeckling of symptomatic images in which the expert is interested in identifying the degree of stenosis and the plaque borders. The proper selection of a despeckle filter is very important in the enhancement of ultrasonic imaging of the carotid artery. Further work is needed to evaluate at a larger scale and in clinical practice the performance of the proposed despeckle filters in the automated segmentation, texture analysis, and classification of carotid ultrasound imaging.     

Wavelet packet correlation methods in biometrics

We introduce wavelet packet correlation filter classifiers. Correlation filters are traditionally designed in the image domain by minimization of some criterion function of the image training set. Instead, we perform classification in wavelet spaces that have training set representations that provide better solutions to the optimization problem in the filter design. We propose a pruning algorithm to find these wavelet spaces by using a correlation energy cost function, and we describe a match score fusion algorithm for applying the filters trained across the packet tree. The proposed classification algorithm is suitable for any object-recognition task. We present results by implementing a biometric recognition system that uses the NIST 24 fingerprint database, and show that applying correlation filters in the wavelet domain results in considerable improvement of the standard correlation filter algorithm.     

Fluorescence spectral imaging for characterization of tissue based on multivariate statistical analysis

A novel spectral imaging method for the classification of light-induced autofluorescence spectra based on principal component analysis (PCA), a multivariate statistical analysis technique commonly used for studying the statistical characteristics of spectral data, is proposed and investigated. A set of optical spectral filters related to the diagnostically relevant principal components is proposed to process autofluorescence signals optically and generate principal component score images of the examined tissue simultaneously. A diagnostic image is then formed on the basis of an algorithm that relates the principal component scores to tissue pathology. With autofluorescence spectral data collected from nasopharyngeal tissue in vivo, a set of principal component filters was designed to process the autofluorescence signal, and the PCA-based diagnostic algorithms were developed to classify the spectral signal. Simulation results demonstrate that the proposed spectral imaging system can differentiate carcinoma lesions from normal tissue with a sensitivity of 95% and specificity of 93%. The optimal design of principal filters and the optimal selection of PCA-based algorithms were investigated to improve the diagnostic accuracy. The robustness of the spectral imaging method against noise in the autofluorescence signal was studied as well.     

基于谱聚类波段选择的高光谱图像分类

高光谱图像在地物观测领域得到了广泛的应用。由于高光谱图像具有数据量大、波段间相关度高等特性,波段选择技术成为降低地物识别计算复杂度的重要方法。根据不同波段数据之间的非线性关系,提出了基于谱聚类(SC)的波段选择技术。该方法首先以波段图像为样本点生成近邻图和相似度矩阵,然后借助谱聚类方法将所有数据样本分成 k类,从中选择 k个代表波段参与后继的分类识别任务。实验数据表明,新方法减小了计算复杂度,提高了地物识别的精度。