基于图像融合和DCT域增强的人脸识别

提出了一种基于图像融合和DCT域增强的人脸识别方法.首先对人脸图像分别进行直方图均衡化和对指数变换,将二者变换结果融合,减少光照对人脸识别的影响;然后进行DCI变换,在滤除高频分量的同时进行图像增强,抑制块效应,在此基础上进行IDCT图像重建,利用二维主成分分析提取人脸特征,降低空间维数;最后通过最邻近分类法实现人脸识别.Yale人脸库仿真实验表明,该方法在光照变化较大和人脸样本较少的情况下具有较高的识别率.     

Face Hallucination with Weighted Nuclear Norm Constraint

Face hallucination via patch-pairs leaning based methods has been wildly used in the past several years. Some position-patch based face hallucination methods have been proposed to improve the representation power of image patch and obtain the optimal regressive weighted vector. The rationale behind the position-patch based face hallucination is the fact that human face is always highly structured and consequently positioned and it plays an increasingly important role in the reconstruction. However, in the existing position-patch based methods, the probe image patch is usually represented as a linear combination of the corresponding patches of some training images, and the reconstruction residual is usually measured using the vector norm such as 1-norm and 2-norm. Since the vector norms neglect two-dimensional structures inside the residual, the final reconstruction performance is not very satisfactory. To cope with this problem, we present a weighted nuclear-norm constrained sparse coding (WNCSC) model for position-patch based face hallucination. In addition, an efficient algorithm for the WNCSC is developed using the alternating direction method of multipliers (ADMM) and the method of augmented Lagrange multipliers (ALM). The advantages of the proposed model are twofold: in order to fully make use of low-rank structure information of the reconstruction residual, the weighted nuclear norm is applied to measure the residual matrix, which is able to alleviate the bias between input patches and training data, and it is more robust than the Euclidean distance (2-norm); the more flexible selection method for rank components can determine the optimal combination weights and adaptively choose the relevant and nearest hallucinated neighbors. Finally, experimental results prove that the proposed method outperforms the related state-of-the-art methods in both quantitative and visual comparisons.     

路面破损图像自动处理技术研究进展

总结了路面破损图像自动处理技术的重要研究成果, 分析了该领域关键技术的研究进展, 包括路面破损检测系统、图像处理算法和识别算法评估; 比较了不同路面破损检测系统与目标自动识别算法的检测精度和适用性, 给出了路面破损图像自动处理技术未来可能的主要研究方向。研究结果表明: 在路面破损检测系统方面, 从早期基于摄影技术的图像采集到目前的3D激光扫描技术, 路面图像采集技术更加便捷和高效, 但破损图像自动分析和目标自动识别算法仍然存在挑战; 在路面破损图像处理算法方面, 传统的路面裂缝目标分割算法已由过去的基于单一特征(灰度、边缘形状等) 的检测方法演化到多特征融合检测方法和图优化检测方法, 还出现了一些精细化的裂缝目标连接与恢复算法, 大幅提高了裂缝检测精度, 但需要的计算资源和人工先验知识库也随之不断增大; 在路面裂缝处理算法评估和比较方面, 主要利用人工分割来评价自动识别结果, 目前迫切需要建立一个面向全球开放的大型路面破损图像数据库, 以客观、有效地评估现有各种路面破损图像处理算法; 基于2D图像特征分析的路面破损图像自动识别算法很难在识别精确性、算法通用性和实时性方面同时取得最佳效果; 近年来, 大量学者开始尝试借助深度学习神经网络自动识别路面破损, 但该技术仍处于活跃的演进过程中; 在提高路面破损自动识别精度和效率方面, 3D激光扫描技术和基于人工智能的深度学习技术的发展将对未来路面破损图像自动识别技术的最终突破产生重大推进作用。      

COLOR IMAGE QUANTIZATION WITH EVOLUTIONARY PROGRAMMING

Introduction　　 The need for quantization of color images isarised because of the limitations of image displayand hardcopy,data storage and data transmissiondevices.The color image quantization is a complexdata clustering problem due to the broad distribu-tion of local optima in the three- dimensional colorspace. Many of the present algorithms for colorquantization find non- optimal solutions,giving riseto visible shifts in color and false contours whenthe number of quantization colors is sma…     

Sequential similarity detection algorithm based on image edge feature

Abstract： This paper proposes a new sequential similarity detection algorithm （SSDA）, which can overcome matching error caused by grayscale distortion; meanwhile, time consumption is much less than that of regular algorithms based on image feature. The algorithm adopts Sobel operator to deal with subgraph and template image, and regards the region which has maximum relevance as final result. In order to solve time-consuming problem existing in original algorithm, a coarse-to-fine matching method is put forward. Besides, the location correlation keeps updating and remains the minimum value in the whole scanning process, which can significantly decrease time consumption. Experiments show that the algorithm proposed in this article can not only overcome gray distortion, but also ensure accuracy. Time consumption is at least one time orders of magnitude shorter than that of primal algorithm.     

Single Image Super-Resolution Method via Refined Local Learning

In this paper,we propose a refined local learning scheme to reconstruct a high resolution(HR)face image from a low resolution(LR)observation.The contribution of this work is twofold.Firstly,multi-direction gradient features are extracted to search the nearest neighbors for each image patch,then the non-negative matrix factorization(NMF)is used to reduce the complexity in weight calculation,and the initial HR embedding is estimated from the training pairs by preserving local geometry.Secondly,a global reconstruction constraint and post-processing by non-local filtering is incorporated into super-resolution(SR)reconstruction process to reduce the image artifacts and further improve the image visual quality.Experimental results show that the proposed algorithm improves the SR performance both in subjective and objective assessments compared with several existing methods.     

A POCS-Based Algorithm for Blocking Artifacts Reduction

Introduction Block-based discrete cosine transform(BDCT) is considered to be a basic tool in bothstill and moving image compression standards,such as JPEG, H.261/263, and MPEG, owing toits near-optimal energy compaction property andmoderate hardware complexity. However, sinceeach N×N block is coded independently, the re-constructed images would exhibit annoying discon-tinuities along block boundaries at low bit rate,which are often referred to as the“blocking arti-facts”in the transfo…     

基于血流图双树复数小波域傅里叶变换的红外人脸识别方法

提出了一种基于血流图与双树复数小波域傅里叶变换的红外人脸识别方法．首先利用血流模型把温谱图转换成血流图,然后将人脸血流图进行一级双树复数小波分解,保留分解后的4个低频子带并分别进行傅里叶变换,得到每个低频子带的特征矩阵,分别计算4个子带的欧氏距离并进行简单的加法融合,再用三阶近邻分类器得到最终的识别结果．为了减小算法的时间复杂度,我们对血流模型进行简化．实验结果表明,本文所提的方法有较好的识别结果．而简化的血流图相比原模型的识别率没有明显的下降,甚至某些情况下还稍高于血流模型,说明本文的方法是有效的．     

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.     

基于APCENN与RBFNN的人脸识别方法研究

提出了将自适应主分量提取神经网络(APCENN)与径向基神经网络(RBFNN)结合进行人脸识别的方法.由于人脸图像维数高,传统主分量分析方法提取人脸主分量运算复杂、速度慢,应用APCENN通过并行运算直接提取人脸主分量,提高了特征提取速度.再通过RBFNN进行识别分类,实验证明网络训练收敛速度快、识别率高.     

低可见度环境下基于同步定位与构图的无人驾驶汽车定位算法

为在大范围低可见度环境下实现无人驾驶汽车的高精度定位，基于VINS-Mono算法的系统框架，在系统的前端与后端分别增添了RFAST弱光图像增强模块与VG融合定位模块，提出了一种融合定位算法LVG_SLAM; RFAST弱光图像增强模块采用小波变换将原始输入图像的细节信息与亮度信息分离，对于包含原始图像噪声的细节信息通过统一阈值和均值滤波2种方式实现噪声抑制，并利用双边纹理滤波算法进行细节增强，在此基础上，根据多尺度Retinex算法增强图像的对比度，提高低可见度环境下角点提取的成功率，从而保证图像跟踪的稳定性，改善定位算法的鲁棒性; 基于无迹卡尔曼滤波算法，VG融合定位模块将GNSS定位信息与惯性导航测量信息进行松耦合，融合定位结果作为约束引入VI-SLAM后端，通过联合非线性优化的方式减少累积误差对算法定位精度的影响。计算结果表明:相较于VINS-Mono算法，改进的LVG_SLAM融合定位算法在EuRoC与Kitti公开数据集上表现更加出色，均方根误差分别降低了38.76%与58.39%，运动轨迹更贴近真实轨迹; 在实际夜晚道路场景下，LVG_SLAM算法将定位误差控制在一定范围内，顺利检测到闭环使得定位表现得到大幅改善，均方根误差、平均误差、最大误差、中位数误差分别降低了79.61%、82.50%、71.31%、83.77%，与VINS-Mono算法相比，在定位精度与鲁棒性方面具有明显的优势。      

基于深度学习的公路货车车型识别

为判断公路货车车型，并提升货车车型识别的速度与精度，提出基于深度学习的方法对公路货车及其轮轴进行精细化目标检测；采用道路监控拍摄和网络爬取的方式获得了16 403张公路货车侧方图像，建立了货车侧方图像数据集，并采用Retinex理论和加入限制对比度的自适应直方图均衡化(CLAHE)等视觉增强方法预处理所采集图像中的光照不均图像和夜视图像；通过理论分析和对比试验选取单阶段检测网络YOLOv3作为公路货车车型识别的目标检测网络，并从调整先验框和模型输入大小以及引入注意力机制3个方面优化了检测模型；针对单帧图像可能同时出现多辆货车的情况，采用基于目标位置信息挖掘的算法分析了货车与轮轴的位置信息，提出一种通过轮轴中心点与货车预测框位置信息判定公路货车与轮轴隶属关系的方法。研究结果表明：图像经过预处理可显著增强车辆的特征信息，优化后检测模型的网络性能得到提高，通过对目标位置信息的挖掘与利用可以很好地解决货车车型判定问题；优化后的检测模型实时检测速度可达47帧·s^-1,对公路货车车型的识别综合准确率达到了94.4%。该方法实现了对公路货车车型的无接触、快速和准确识别，为公路货...     

基于改进YCrCb颜色空间的肤色分割

人脸识别是计算机视觉、模式识别、生物特征识别、图像处理等多学科交叉的一门科学,肤色分割是人脸识别研究中一个重要的步骤。针对基于YCrCb颜色空间的肤色分割在强光下分割不准确的现状,提出了一种基于改进的YCrCb空间的肤色分割方法,本文利用改进YCrCb颜色空间提高了强光下肤色分割的准确性和适应性。实验证明该算法减弱了光线强弱对人脸肤色分割的影响,更加准确的找到人脸区域,误检率较低。     

数字图像增强算法分析

在阐明图像增强处理基本方法基础上,对几种有代表性的图像增强算法(基于直方图均衡化图像增强算法,基于模糊集理论的图像增强算法,基于小波变换的图像增强算法,基于人眼视觉特性的彩色图像增强算法及基于神经网络的图像增强算法)做简单介绍,对现有直方图的均衡化算法进行分析、对比,综合多种算法对现有直方图均衡化算法做改进,得出一种新的直方图均衡化方法。     

基于半监督哈希算法的交叉口交通状态识别

准确辨识交叉口交通状态是实施有效交通控制策略的前提. 传统交通状态识别方法是利用占有率、排队等统计数据设计指标实现状态识别，存在只能从单一角度刻画交叉口交通需求的问题. 对此，提出基于半监督哈希算法的交叉口交通状态识别方法. 从原始数据丰富特征入手，构建交叉口有效检测区域的图像化模型；将交叉口交通状态识别转化为图像搜索问题，利用监督哈希算法实现基于部分标签信息的图像搜索，进而得到交叉口的交通状态；最后，利用仿真对该方法进行了验证. 结果表明，所提方法在识别精度和速度上具有可行性和有效性.     

O形密封圈尺寸视觉精密测量技术

针对现有O形密封圈尺寸测量方法的局限性,设计基于主动视觉的精密检测方法,并对快速自动调焦、亚像素边缘检测、边缘点拟合技术进行了研究。将快速自动调焦过程分为3个阶段,使用大步长和方差函数,结合灰度切割对比度增强与中值滤波,完成粗调焦;使用小步长和Tenengrad函数,完成细调焦;采用Brenner函数和微步距完成高灵敏度调焦.用Zernike矩算法对边缘点进行亚像素定位,并采用面积最小误差法拟合边缘点,求取O形密封圈的内径、外径.实验结果表明:本文方法经过16步搜索,在15 s内可自动完成调焦;与传统方法相比,通过亚像素提取,能将内外径测量结果达到微米级,满足O形密封圈测量需求.      

DIRECT VOXEL-PROJECTION FOR VOLUMETRIC DATA RENDERING IN MEDICAL IMAGERY

IntroductionWith the development of medical imagingtechniques,many devices such as computed tomo-graphy,positron emission tomography,singlephoton emissioin computed tomography,magneticresonance imaging,and ultrasound scanners arewidely available for capturing the planar images ofhuman organs directly in machine- readable form.These medical imaging methods bring a lot of ad-vantages to clinical diagnosis and treatment.How-ever,due to these systems'2 - D imaging modali-ties,it is difficult to un…     

基于高斯混合部件模型的铁路扣件检测

针对采集图像中铁路扣件存在形状的变化、扣件图像的光照差异较大和扣件被异物局部遮挡的问题,根据对可变形部件模型算法和高斯混合模型的研究,提出了高斯混合部件模型算法. 结合扣件图像边缘特性及改进的Roberts算子计算图像梯度,将归一化后的方向梯度直方图特征作为高斯混合部件模型算法的底层特征,根据扣件形状划分部件,部件之间的相对位置采用星型连接方式度量,运用余弦相似性度量部件中方向梯度直方图特征的相似度,部件模型使用高斯混合模型并采用期望最大化算法迭代求解. 将高斯混合部件模型算法应用于扣件检测中,最终平均检测效果为漏检率3.16%、误检率9.80%、正确率90.27%.      

多层感知器自监督在线修正的道路识别算法

为解决自主移动机器人非结构化道路识别检测准确性、鲁棒性及实时性的问题，提出一种基于感兴趣区域(Region of Interest，ROI)与多层感知器(Multi-Layer Perceptron，MLP)为核心的自监督在线修正算法.首先，通过ROI算法规定被处理图像的有效计算区域；其次，利用多层感知器对样本数据进行训练，将感兴趣区域按相应特征实现分类处理，并对分类区域进行形态学处理及特征提取处理，筛选出有效的行驶区域；最后，通过自监督在线修正算法替换错误处理结果，进一步保障道路分类识别的准确性.实验结果表明，改进算法能准确地识别出环境中的道路区域，具有良好的实时性与可靠性.     

多层感知器自监督在线修正的道路识别算法

为解决自主移动机器人非结构化道路识别检测准确性、鲁棒性及实时性的问题,提出一种基于感兴趣区域(Region of Interest,ROI)与多层感知器(Multi-Layer Perceptron,MLP)为核心的自监督在线修正算法.首先,通过ROI算法规定被处理图像的有效计算区域;其次,利用多层感知器对样本数据进行训练,将感兴趣区域按相应特征实现分类处理,并对分类区域进行形态学处理及特征提取处理,筛选出有效的行驶区域;最后,通过自监督在线修正算法替换错误处理结果,进一步保障道路分类识别的准确性.实验结果表明,改进算法能准确地识别出环境中的道路区域,具有良好的实时性与可靠性.