Iris Localization Based on Edge Searching Strategies

An iris localization scheme based on edge searching strategies is presented. First, the edge detection operator Laplacian-of-Gaussian （LOG） is used to iris original image to search its inner boundary. Then, a circle detection operator is introduced to locate the outer boundary and its center, which is invariant of translation, rotation and scale. Finally, the method of curve fitting is developed in localization of eyelid. The performance of the proposed method is tested with 756 iris images from 108 different classes in CASIA Iris Database and compared with the conventional Hough transform method. The experimental results show that without loss of localization accuracy, the proposed iris localization algorithm is apparendy faster than Hough transform.     

云遗传算法

为了克服传统遗传算法搜索速度慢、易陷入局部最优解的缺陷，借鉴遗传算法的思想，利用云模型云滴的随机性和稳定倾向性的特点，提出了一种新的遗传算法——云遗传算法（CGA）．该算法由正态云模型的Y条件云发生器实现交叉操作，由基本云发生器实现变异操作．最后，进行了函数优化实验，并与标准遗传算法（SGA）和自适应遗传算法（AGA）进行了比较，以证明其有效性．     

Efficient iris recognition via ICA feature and SVM classifier

With the increasing demands of security in ournet worked society,the technology for personalidentification works as the main solution to safe-guard people s properties.Biometrics is an alterna-tive to solve the problemand has the advantage thatthey cannot be stolen or forgotten like pass words.Because personal identification numbers or identifi-cation tokens(such as s mart cards)cannot provide ahighlevel of security which can be copied,inspec-ted and/or stolen.They only showthe knowledgeof som…     

NEW HAUSDORFF DISTANCE-BASED VEHICLE LOCATION

Objective To realize accurate localization of moving vehicles from single monocular intensity image.Methods The new modified Hansdorff distance(M2HD) was adopted, which used dominant points instead of edge maps as features for measuring similarity between image and model projection. Modified simulated annealing (MSA) algorithm was used to find optimum localization parameters. Results M2HD reduces the computational complexity,and improves the matching precision. Furthermore, MSA can fast find global optimum instead of getting into partial one because of its high parallel and robust performance. Conclusion Experiments confirm that the combination of MSA and M^2 HD can effectively localize the vehicles that are changed both in translation and rotation     

基于多层前馈神经网络的车牌提取方法

文中提出一种采用三层前馈神经网络提取车牌的方法。该方法是一种基于纹理分析的图像分割算法,适合于彩色及灰度图像。实验表明,该方法能准确地提取图像中的车牌。     

基于RSSI的无线传感器网络加权质心定位算法

节点定位是无线传感器网络中的关键技术之一．文中通过对无线电传播路径损耗模型的分析，提出了加权质心定位算法，用信标节点对未知节点的不同影响力来确定加权因子，以提高定位精度．并且在理论分析的基础上，提出了优选信标节点进行节点定位计算的规则，以此进一步提高节点定位精度．加权质心定位算法计算简单，定位过程中节点问无通信开销．节点定位精度较常用的极大似然估计算法高，具有较普遍的应用意义．     

Eye and Iris Detection Using Projection and Radial Symmetry Transform

This paper presents an eye and iris detection algorithm for human facial images. The authors combine three features of the eye to develop the algorithm: 1) the pixels surrounding the eyes are more variable than other parts of the face; 2) eye pixels are darker than their neighbors; 3) eyes often exhibit radial symmetric properties. Through the first feature, two rough regions of both eyes are detected on the face. Eye masks are then formed based on the second feature, and a fast radial symmetry transform is applied to the two rough regions of both eyes. Finally, accurate iris centers are located by searching the maximum value of the radial symmetry transform results. Using 450 human facial images from the Caltech face database, experiments show that the success rate of the proposed method is 91.7%. The effectiveness of the method was also verified through detection of video frames.     

应用形态学方法提高SAR目标检测中的聚类效果

Lincoln实验室提出的SAR(synthetic aperture radar)ATR(automatic target recognition)算法由于其经典性而被广泛采用,该算法为三级结构模式,处于其中Prescreener级和Discriminator级之间的聚类算法对于检测算法整体性能有重要的影响,文中介绍了SAR ATR算法采用的常规聚类算法,分析了常规算法在聚类过程中存在的杂波干扰问题,针对问题在聚类前引入形态学操作方法,将待聚类图像中包含的孤立点删除而只保留团状分布的样本,从而消除了杂波点对聚类的干扰,基于实际SAR图像的聚类结果验证了应用形态学方法对提高聚类效果的有效性.     

无线传感器网络定位算法的研究

在无线传感器网络中，因无线传感器设备功耗、价格和硬件限制及对定位精度的要求，距离无关定位机制被认为是一类具有成本效益的解决方案。介绍了距离无关定位机制中的DV—hop定位算法，研究了不良节点、网络拓扑结构、锚节点和邻居节点对DV—hop定位算法性能的影响，分析了算法的通信量和计算量。在连通的网络中，该算法能够达到35％～40％的平均定位精度；该算法只需要较少的锚节点，计算和通信开销适中，不需要节点具备测距能力，是一个可扩展的算法，适用于各向同性的密集网络。     

Novel Chaotic Fragile Watermarking Algorithm for Image Authentication

Taking full advantage of the randomicity of chaotic system and its extreme sensitivity to the initial value, a new chaotic fragile watermarking algorithm is proposed. In the algorithm, the location key is looked as a logistic chaotic initial value for iteration to create a location matrix. According to this location matrix, a mapping image is generated and the embedding location of watermarking in image blocks is identified. Then, the watermarking sequence, which is related with the mapping image blocks and generated by H6non chaotic map, is embedded into the least significant bit (LSB) of the corresponding location in each block. Since the image content and watermarking are staggered, the algorithm has a higher security. Simulation results showed that the algorithm can detect and locate the tamper in watermarked images with an accuracy of 2 x 2 block pixels. At the same time, the watermarking images has good invisibility, and the original image is not required when extracting watermarking.     

基于量子遗传优化算法的图像稀疏分解

为了减少图像稀疏分解的计算量,提出了一种基于量子遗传算法与匹配追踪相结合的图像稀疏分解快速算法.量子遗传算法能用较小的种群规模实现较大的空间搜索,全局寻优能力强,基于匹配追踪的图像稀疏分解是最优化问题,因此可用量子遗传算法快速实现.仿真结果表明,每步分解所需计算的图像或图像残差与原子的内积仅4 000次,由分解结果重建的图像具有较好的主观质量.     

求解Job-Shop车间作业调度的混合算法

提出了一种解决车间作业调度最短完成时间的有效的混合算法．基于考虑到遗传算法的早熟收敛问题和禁忌搜索算法的自适应优点，该算法结合了遗传算法和禁忌搜索算法对Job—Shop车间作业调度问题的解进行编码，通过实例表明该算法是可行和有效的．     

A generalized two-level Bregman method with dictionary updating for non-convex magnetic resonance imaging reconstruction

In recent years, it has shown that a generalized thresholding algorithm is useful for inverse problems with sparsity constraints. The generalized thresholding minimizes the non-convex p-norm based function with p 1, and it penalizes small coefficients over a wider range meanwhile applies less bias to the larger coefficients.In this work, on the basis of two-level Bregman method with dictionary updating(TBMDU), we use the modified thresholding to minimize the non-convex function and propose the generalized TBMDU(GTBMDU) algorithm.The experimental results on magnetic resonance(MR) image simulations and real MR data, under a variety of sampling trajectories and acceleration factors, consistently demonstrate that the proposed algorithm can efficiently reconstruct the MR images and present advantages over the previous soft thresholding approaches.     

一种改进的遗传算法及其在作业车间调度的应用

基于遗传算法容易产生早熟的现象,提出了一个概念清楚、运算量小的评价种群过早收敛程度的指标,并利用该指标提出了一种新的调整策略.仿真实例表明,该方法能及时反映种群在进化过程中的过早收敛程度,不仅能加快计算速度,而且还能增强算法的全局收敛性.     

基于路面指纹的高精度车辆定位

车辆定位广泛使用基于视觉的定位方法,针对前视图像或侧视图像易受到周围环境的影响且定位过程中需要遍历匹配地图图像导致耗时较长的问题,本文提出一种基于俯视路面图像的表征模型--路面指纹.路面指纹包含GPS(Global Positioning System),路面特征和图像特征点.该模型通过卷积神经网络(Convolutional Neural Network, CNN)结合连通区域识别待定位图像的路面特征信息,利用路面特征信息对GPS初定位筛选的地图节点进一步筛选从而提高定位效率.分别在路面特征信息密集和稀疏的路段下进行实验,实验结果表明,通过引入路面指纹使定位耗时减少20.3%,平均定位误差为47.4 mm.该方法能提高定位效率并实现高精度车辆定位.     

高铁扣件的自适应视觉检测算法

为了实现高铁缺陷扣件的准确、快速和自动化检测，提出一种基于图像处理技术的高铁扣件自适应视觉检测算法. 针对高铁扣件图像的特性，使用改进的LBP (local binary pattern)算子提取扣件的显著特征；在扣件特征图的基础上，采用模板匹配算法得到扣件区域在原始图中的精确位置，进而得到扣件子图并用扣件的位置信息校验定位结果；以相邻两个扣件子图的差值作为判断依据，如果差值大于预设的阈值，相应的扣件则被判断为缺陷扣件. 将该检测算法应用于高铁工务部门提供的真实扣件图. 研究结果表明:本文提出的自适应扣件检测算法在雨天的表现最差，检出率为96%，误检率为0.50%；在晴天的表现最好，检出率为100%，误检率为0.22%；在不同天气、光照、环境下的综合检出率为99%，综合误检率为0.33%.      

MEDICAL IMAGE SEGMENTATION BASED ON A MODIFIED LEVEL SET ALGORITHM

Imagesegmentationplaysanimportantrolein imageanalysisaswellasinhigh levelimageinter pretationandunderstandingsuchasrobotvision, objectrecognition,andmedicalimaging.Numerous segmentationmethodshavebeendeveloped.How ever,medicalimagesareoftencorruptedbyno…     

Exploiting Local and Global Characteristics for Contrast Based Visual Saliency Detection

Visual saliency is an important cue in human visual system to identify salient region in the image;it can be useful in many applications including image retrieval,object recognition,image segmentation,etc.Image contrast has been used as an effective feature to detect visual salient region.However,the conventional contrast measures either in spectral domain or in spatial domain fail to give sufficient consideration towards the local and global characteristics of the image.This paper presents a visual saliency detection algorithm based on a novel contrast measurement.This measurement extracts the spectral information of image block using the 2D discrete Fourier transform(DFT),and combines with the total variation(TV)of image block in spatial domain.The proposed algorithm is used to perform salient region detection in the image,and compared with state-of-the-art algorithms.The experimental results from the MSRA dataset validate the effectiveness of the proposed algorithm.     

一种考虑分类的石材图像相似性比对算法

在基于颜色的图像相似比较中,传统的处理方法大多采用各子空间的概率作为特征,在此基础上定义图像的距离和相似度.但这些方法可能导致色彩比对失真的情况出现;在同一类的石材中颜色和纹理有很大的差异,甚至会出现两张分属异类的石材图像在单纯从颜色、纹理等特征上进行判断比同类很多石材图像更为"相似".文中引入了先对石材的样本图像进行分类,定义图像与各大类的相似度.在类内定义较为一般性的相似度.综合二者给出图像的整体相似度定义,实验表明该方法具有较好的分类效果和较合理的石材图像相似度.     

车辆定位的地图匹配算法研究

车载导航定位系统是智能交通系统的重要组成部分,其中地图匹配又是提高车辆定位精度的重要技术。从数字地图本身的属性特征出发,利用数字地图的地面高程模型对现有的地图匹配算法进行改进,并通过实验数据对改进后的算法加以验证,其结果表明改进后的算法能够获得较高的定位精度和更可靠的匹配性能。