基于改进SIFT特征和图转换匹配的图像匹配算法

针对SIFT特征在纹理丰富的图像中提取较多的伪点和不稳定的点而影响图像匹配的问题, 提出了一种基于Harris阈值准则的局部不变特征图像匹配算法。该算法在提取SIFT不变特征的基础上, 利用Harris阈值准则对所提取到的不变特征进行选择, 剔除了图像区域中大量可区分性较差的特征点, 从而得到了相对稳定和可区分性较好的特征点。其次, 结合不变特征矢量与图转换匹配(GTM)的方法对提取到的稳定特征点进行了精确匹配。实验对比结果表明, 用取得稳定的特征点, 进而结合一种好的匹配策略, 能够更加增强图像匹配的高效性和鲁棒性。     

A novel multiset integrated canonical correlation analysis framework and its application in feature fusion

Multiset canonical correlation analysis (MCCA) is difficult to effectively express the integrated correlation among multiple feature vectors in feature fusion. Thus, this paper firstly presents a novel multiset integrated canonical correlation analysis (MICCA) framework. The MICCA establishes a discriminant correlation criterion function of multi-group variables based on generalized correlation coefficient. The criterion function can clearly depict the integrated correlation among multiple feature vectors. Then the paper presents a multiple feature fusion theory and algorithm using the MICCA method. The detailed process of the algorithm is as follows: firstly, extract multiple feature vectors from the same patterns by using different feature extraction methods; then extract multiset integrated canonical correlation features using MICCA; finally form effective discriminant feature vectors through two given feature fusion strategies for pattern classification. The multi-group feature fusion method based on MICCA not only achieves the aim of feature fusion, but also removes the redundancy between features. The experiment results on CENPARMI handwritten Arabic numerals and UCI multiple features database show that the MICCA method has better recognition rates and robustness than the fusion methods based on canonical correlation analysis (CCA) and MCCA.     

区域加权信息熵及其在图像特征提取中的应用

图像特征提取是当前基于内容图像检索领域的研究重点,然而单纯基于信息熵的图像特征提取方法无法体现图像内容的位置信息.分析现有的基于颜色-空间图像特征提取算法的基础上,结合图像信息熵概念与图像分割算法,提出了一种新的图像信息熵描述方法,即区域加权信息熵,并证明了区域加权信息熵的若干性质.采用信息熵性能评价指标从概率的角度描述因权值变化而引起的图像信息熵分布的变化,并考虑应用的兴趣区域以及权值粒度从而确定合理权值.实验表明区域加权信息熵方法比单纯信息熵方法描述图像内容准确率提高了50%以上.     

有限制的MFA-ICA的算法及其在图像特征提取中的应用

统的独立成分分析(IndependentComponentAnalysis,ICA)是一种无噪声模型,而实际应用中噪声是存在的。根据多元统计中的因子分析模型,改变其假设条件,从而得到一种有噪声ICA模型,对于模型参数,引入平均场近似(MeanFieldApproximation,MFA)原理来求解。针对图像特征提取,通过增加对模型参数的一些限制,使其能得到更为独立的图像特征,为图像识别提供更可靠的特征信息,从而大大提高识别率。通过仿真模拟图形以及ORL人脸数据进行实验,将传统的独立成分分析算法、无限制的MFA ICA算法以及增加限制条件的MFA ICA算法进行比较,从仿真模拟图形实验结果看,限制的MFA ICA算法能分离出更独立的特征,同时利用限制的MFA ICA算法识别效果明显优于传统ICA算法和无限制MFA ICA算法。     

二维典型相关分析在小样本图像识别上的应用

针对传统典型相关分析（Canonical Correlation Analysis,CCA）的图像识别中出现的小样本（Small Sample Size,SSS）问题,提出二维典型相关分析（Two-Dimensional CCA,2DCCA）。首先阐述了2DCCA方法的基本原理并给出了类成员关系矩阵的构造方法,推导出了类成员关系协方差矩阵广义逆的解析解。其次,从理论上证明了2DCCA方法对于解决小样本问题的有效性。最后,利用人脸识别实验来测试该方法的性能,实验结果表明,2DCCA方法有效地解决了图像识别中常见的小样本问题,并且能取得较其他几种基于CCA的人脸识别方法更优的识别结果。     

离散脊波变换及其在SAR图像压缩中的应用

首先介绍了SAR图像的特点，在对离散脊波变换的正交性及其与Radon变换的关系深入研究的基础上，利用其在线性特征表示方面的优势，将离散脊波变换应用到SAR图像的压缩，提出了一种基于离散正交脊波变换的SAR图像压缩方法。实验数据表明该算法对斑点噪声污染严重的SAR雷达图像压缩较传统的小波变换更加有效。     

Improving pilot mental workload classification through feature exploitation and combination: a feasibility study

Predicting high pilot mental workload is important to the United States Air Force because lives and aircraft have been lost due to errors made during periods of flight associated with mental overload and task saturation. Current research efforts use psychophysiological measures such as electroencephalography (EEG), cardiac, ocular, and respiration measures in an attempt to identify and predict mental workload levels. Existing classification methods successfully classify pilot mental workload using flight data for a single pilot on a given day, but are unsuccessful across different pilots and/or days. We demonstrate a small subset of combined and calibrated psychophysiological features collected from a single pilot on a given day that accurately classifies mental workload for a separate pilot on a different day. We achieve classification accuracy (CA) improvements over previous classifiers exceeding 80% while using significantly fewer features and dramatically reducing the CA variance. Without the need for EEG data, our feature combination and calibration scheme also radically reduces the raw data collection requirements, making data collection immensely easier to manage and spectacularly reducing computational processing requirements.     

面向彩色图像的尺度和旋转不变性特征提取方法及应用

提出一种面向彩色图像的尺度和旋转不变性特征提取方法,并在真实的场景识别中进行了应用。该方法是先对给定彩色图像的各组成平面分别进行Radon变换,然后对得到的Radon变换系数矩阵进行尺度不变性处理,接着对处理后的Radon变换系数矩阵用频率B样条小波进行1维小波变换,在所得到的脊波系数矩阵中计算均值和方差的同时,采用线性回归模型提取在不同的颜色组成平面下所有频率子波段之间的关系属性,最后将得到的特征进行旋转不变性处理,从而得到所提出的尺度和旋转不变性特征。在3个数据库上进行了实验,结果表明本文方法可靠有效。     

FD-TR: feature detector based on scale invariant feature transform and bidirectional feature regionalization for digital image watermarking

Multimedia Tools and Applications - In this paper we propose the FD-TR: Feature Detector Based on Scale Invariant Feature Transform and Bidirectional Feature Regionalization for digital image...     

RETRACTED ARTICLE: Enhanced artificial bee Colony algorithm and its application in multi-threshold image feature retrieval

Multimedia Tools and Applications - In human information, visual information accounts for about 60%. Image is the main way for humans to obtain visual information. Image binarization is an...     

基于颜色常量的图像特征提取及其降维方法

在图像特征提取过程中,为了提高图像检索中的抗光照影响能力和检索效率,给出了图像颜色索引提取的一种方法-颜色常量方法,利用该方法能减少光照强度变化对颜色检索的影响;给出了图像检索中的一种颜色特征降维方法-主分量分析方法,利用该方法能有效减小特征数据库的大小,提高图像的匹配速度.实验结果表明,给出的方法不但能有效地减少光照强度变化和噪音对图像颜色检索的影响,而且能够有效地提高检索的效率,检索的结果能较好地接近于人们的视觉感知结果.     

Successive pattern classification based on test feature classifier and its application to defect image classification

A novel successive learning algorithm based on a Test Feature Classifier is proposed for efficient handling of sequentially provided training data. The fundamental characteristics of the successive learning are considered. In the learning, after recognition of a set of unknown data by a classifier, they are fed into the classifier in order to obtain a modified performance. An efficient algorithm is proposed for the incremental definition of prime tests which are irreducible combinations of features and capable of classifying training patterns into correct classes. Four strategies for addition of training patterns are investigated with respect to their precision and performance using real pattern data. A real-world problem of classification of defects on wafer images has been dealt with by the proposed classifier, obtaining excellent performance even through efficient addition strategies.     

Stacked Euler vector (SERVE): a gray-tone image feature based on bit-plane augmentation

A new combinatorial feature called stacked Euler vector (SERVE) is introduced to characterize a gray-tone image. SERVE comprises a four-tuple, where each element is an integer representing the Euler number of the partial binary image formed by certain pixel overlap relations among the four most significant bit planes of the gray-tone image. Computation of SERVE is simple, fast, and does not involve any floating point operation. SERVE can be used to augment other features to improve the performance of image retrieval significantly. Experimental results on the COIL database are reported to demonstrate its performance     

高温环境中线阵CCD图像特点识别算法适应性研究

利用激光三角折射法在高温环境(高于650℃)条件下,对玻璃厚度进行在线测量,要求精度优于±0.02 mm.由于实际测量环境与标准实验室测量环境差距大,能否在恶劣环境中实现像点精确定位是仪器设计的关键问题.在分析高温环境中CCD目标像点热致光学误差、激光传输湍流效应的基础上,利用目标像点内像素间灰度值的高度相关性,构造一个尽可能精确反映目标区域像素灰度值、目标像点质心位置之间关系的数学模型,从而实现对点目标像点位置的精确估计.     

U-Net网络医学图像分割应用综述

病灶精确分割对患者病情评估和治疗方案制定有重要意义,由于医学图像中病灶与周围组织的对比度低,同一疾病病灶边缘和形状存在很大差异,从而增加了分割难度。U-Net是近些年深度学习研究中的热点,为医生提供了一致性的量化病灶方法,一定程度上提高了分割性能,广泛应用于医学图像语义分割领域。本文对U-Net网络进行全面综述。阐述U-Net网络的基本结构和工作原理;从编码器个数、多个U-Net级联、与U-Net结合的其他模型以及3D U-Net等方面对U-Net网络模型的改进进行总结;从卷积操作、下采样操作、上采样操作、跳跃连接、模型优化策略和数据增强等方面对U-Net网络结构改进进行总结;从残差思想、密集思想、注意力机制和多机制组合等方面对U-Net的改进机制进行总结;对U-Net网络未来的发展方向进行展望。本文对U-Net网络的原理、结构和模型进行详细总结,对U-Net网络的发展具有一定积极意义。     

Significance of tectonics in linear feature detection and interpretation on satellite images

There are two main concepts of linear feature evaluation based on satellite images. One puts emphasis on finding equivalence between the image and the field data, in the other, the basic aim is to correlate them after their more or less separate interpretation. A comparative study of the two methods has been carried out on a test field in Southwest Hungary using several kinds of geological and geophysical data as reference. Combining the statistical and deterministic approach a block-faulting mechanism can be recognized and described with ENE-WSW main axis. Correlation of satellite and reference data proved the tectonic difference of the northern and southern parts of the test field. This led to the conclusion that one can derive textural information on tectonics from a Landsat image concerning especially short lines. This implies that not only the present surface expression of faults and tectonics is detectable but the determination of the latest tectonically active surface is also possible.     

基于肤色和类Harr特征的人脸图像的人眼检测

人眼检测在表情识别和计算机视觉领域得到了广泛的关注和研究,但是在多数的人眼检测方法中,对于背景较复杂的图像,识别率急速下降,误检率急剧上升。经过研究,使用椭圆肤色模型预处理图像,分割出肤色区域和非肤色区域,检测算法只对肤色区域进行人眼检测,有效降低了复杂背景造成的高误检率。同时特征选取是决定检测算法识别率和误检率等性能标准的关键因素,选取类Harr特征训练Adaboost级联分类器,实验表明了类Harr特征的有效性。     

多尺度变换像素级医学图像融合：研究进展、应用和挑战

基于多尺度变换的像素级图像融合是计算机视觉领域的研究热点,广泛应用于医学图像处理等领域。本文对多尺度变换的像素级图像融合进行综述,阐述多尺度变换图像融合的基本原理和框架。在多尺度分解方面,以时间为序梳理了塔式分解、小波变换和多尺度几何分析方法的发展历程。在融合规则方面,围绕Piella框架和Zhang框架,讨论通用的像素级图像融合框架;在低频子带融合规则方面,总结基于像素、区域、模糊理论、稀疏表示和聚焦测度的5种融合规则;在高频子带融合规则方面,综述基于像素、边缘、区域、稀疏表示和神经网络的5种融合规则。总结12种跨模态医学图像融合方式,讨论该领域面临的主要挑战,并对未来的发展方向进行展望。本文系统梳理了多尺度变换像素级图像融合过程中的多尺度分解方法和融合规则,以及多尺度变换在医学图像融合中的应用,对多尺度变换像素级医学图像融合方法的研究具有积极的指导意义。     

Eye detection in a facial image under pose variation based on multi-scale iris shape feature

The accurate location of eyes in a facial image is important to many human facial recognition-related applications, and has attracted considerable research interest in computer vision. However, most prevalent methods are based on the frontal pose of the face, where applying them to non-frontal poses can yield erroneous results.In this paper, we propose an eye detection method that can locate the eyes in facial images captured at various head poses. Our proposed method consists of two stages: eye candidate detection and eye candidate verification. In eye candidate detection, eye candidates are obtained by using multi-scale iris shape features and integral image. The size of the iris in face images varies as the head pose changes, and the proposed multi-scale iris shape feature method can detect the eyes in such cases. Since it utilizes the integral image, its computational cost is relatively low. The extracted eye candidates are then verified in the eye candidate verification stage using a support vector machine (SVM) based on the feature-level fusion of a histogram of oriented gradients (HOG) and cell mean intensity features.We tested the performance of the proposed method using the Chinese Academy of Sciences' Pose, Expression, Accessories, and Lighting (CAS-PEAL) database and the Pointing'04 database. The results confirmed the superiority of our method over the conventional Haar-like detector and two hybrid eye detectors under relatively extreme head pose variations.