基于改进遗传算法的视差拼图缝线搜索方法

为了消除视差图像拼接后的局部不一致、不连续或变形,提出一种视差拼图最优合成拼缝搜索策略.将两浇幅图象重叠区域的归一化差图象作为搜索空间,定义一个与图象高相等维数的向量作为染色体,染色体的基因表示每一行图象中的最优拼接点,采用常用的最小值搜索适应度函数作为视差图像拼缝搜索的适应度函数.针对传统遗传算法在图像整数编码搜索中收敛慢的缺陷,提出了一种基于小区间整数编码,个体互异稳态定标的组合选择策略,按基因适应度自适应重组,自适应、变步长与多点选择性的变异操作等策略的改进遗传算法.实际图像拼缝搜索实验表明:改进后算法比基本遗传算法和传统自适应遗传算法收敛速度加快.     

一种基于mean shift的多通道图像分割算法

目的针对目标与背景对象的色彩值比较接近的RGB图像中,目标对象难以有效分割的问题,探索一种基于mean shift的RGB多通道图像的分割方法。方法根据RGB图像的3个通道对颜色的敏感性差异,运用均值偏移算法对RGB图像的3个通道分层聚类,再引入可靠性因子,分别对3个单通道的各聚类像素进行可靠性计算,并保留可靠性高的像素作为分割结果,最后采用逻辑"或"运算融合单通道的分割结果,得到最终分割图像。结果与一般分割算法相比,该方法的分割效果好,误分率低,改善了图像的分割质量。结论该算法具有很好的推广性,能够应用于彩色印品缺陷检测、彩色包装图像检测中。     

基于对象的三维图像颜色传递与视差优化

颜色传递是近年来图像处理和计算机视觉领域的热门研究问题,随着立体图像技术的发展,对于立体图像的颜色传递越来越受关注。本文提出一种双目立体图像的颜色传递方法,在完成颜色传递的同时力求提升用户的观看体验。根据用户实际需求,可以对目标对象进行颜色传递,而保持背景的颜色不改变。在本文提出的方法中,由用户指定图像对象,然后用图割的方法进行图像分割,根据所选对象与目标图像颜色特征的多元高斯模型匹配完成颜色传递。为了进一步增强观看效果,本文在颜色传递的同时进行非线性视差优化,从而提高目标对象的深度感。本文从不同立体图像库中随机选取图像进行实验,实验结果表明,本文方法中颜色传递和视差优化的结合,可以很好地提升立体图像的观看体验。     

基于共轭梯度法的快速Mean Shift图像分割

针对均值漂移算法收敛速度较慢的问题,本文出了基于共轭梯度的快速均值漂移算法,并将其用于图像分割.该算法利用共轭梯度法简便,存储需求小,收敛速度介于最速下降法和牛顿法之间, 具有较好的全局收敛性和较快的收敛速度的特点,通过交替执行均值漂移算法和共轭梯度算法提高经典均值漂移算法的收敛速度.对合成图像和真实图像的实验结果表明了新算法不但提高了经典均值漂移算法的速度,而且在进行图像分割时保持了良好的分割结果.     

基于多特征的模糊c均值聚类的图像分割

图像分割是图像处理中最为基础和重要的领域之一，是对图像进行视觉分析和模式识别的基本前提。本文提出采用像索点在θ方向上的非覆盖均值差分作为像素点的一个重要特征，联合其他常用特征，对图像进行c均值模糊聚类分析。实验结果表明本文提出的方法可以很好的从图像中分割出有意义的区域。     

量子蚁群模糊聚类算法在图像分割中的应用

针对模糊C-均值算法对初始值的依赖,容易陷入局部最优值的缺点,本文提出将量子蚁群算法与FCM聚类算法结合,首先利用量子蚁群算法的全局性和鲁棒性以及快速收敛的优点确定图像的初始聚类中心和聚类个数,再将所得结果作为FCM聚类算法的初始参数,然后用FCM聚类算法对医学图像进行分割。实验结果表明,该方法有效解决了FCM算法对初始参数的依赖,克服了FCM算法及蚁群算法容易陷入局部极值的的缺点,而且在分割速度和精度上得到了较大提高。     

基于FCM和图割的交互式图像分割方法

为了提高在前景和背景颜色相似情况下图像的分割效果,提出了一种基于模糊C均值聚类(FCM)和图割的交互式图像分割方法。首先,利用分水岭算法对图像进行预处理,将图像分成多个小区域,用区域代替像素点进行分析。然后,采用模糊C均值算法对用户标记的前景区域和背景区域分别进行聚类分析,挖掘用户交互所提供的隐藏信息。用未标记区域的颜色分量到前景区域及背景区域类心的最小距离表示相似能量,用未标记区域与其相邻区域的相关性表示先验能量。最后,利用最大流/最小割算法求能量函数的全局最优解。与其他方法相比,该文方法具有较好的分割性能,能从前景背景相似的图像中较精确地提取感兴趣的物体,且用户操作简单。     

模拟退火与模糊C-均值聚类相结合的图像分割算法

模糊C-均值(FCM)聚类算法是一种结合无监督聚类和模糊集合概念的图像分割技术,比较有效,但存在着受初始聚类中心和隶属度矩阵影响,可能收敛到局部极小的缺点.将模拟退火算法(SA)与模糊C-均值聚类算法相结合,在合理选择冷却进度表的基础上,依据模糊C-均值聚类算法建立模拟退火算法的目标函数,实现了基于模拟退火的模糊C-均值聚类图像分割算法.实验表明,该方法具有搜索全局最优解的能力,因而可得到很好的图像分割结果.     

基于SLIC超像素分割的图分割算法

图像分割是对图像进行分析和理解的关键步骤,是计算机视觉的基本技术之一.计算复杂度是评判一个图像分割算法好坏的重要标准,因此降低算法的计算复杂度是当前图像分割领域的主要任务之一.本文提出了一种基于SLIC超像素分割的图像分割方法.该方法利用SLIC算法生成超像素,通过构造相应的相似性矩阵,有效降低了Ncut分割算法的计算复杂度,大幅度缩短了Ncut算法的运行时间.由于SLIC超像素分割算法的准确性与高效性,在进行三类自然图像分割实验时,本文提出的方法无论在分割效果,还是在运行时间上,都要明显优于Ncut分割方法及它的改良算法.     

基于视差图像序列的深度估计算法研究

现有的深度估计算法中,Kim算法不需要全局优化,可以保留精确的目标轮廓,同时仍然确保在少量细节的区域中平滑重建,得到的效果相对较好,但是其在深度扩散算法中时间复杂度太高,不适合广泛应用.针对这一问题,我们对其深度扩散算法进行改进.首先,根据边缘图像的深度范围覆盖整个场景的深度,将深度扩散的方式转变为在边缘深度范围内的快速搜索而非整个区域的全局搜索,明显降低了算法的时间复杂度.另外,根据极平面图像(EPI)的线性结构,将深度扩散沿着斜率的方向对多个视角进行综合考虑扩散,而非只是针对单个视角下的像素点进行,使非边缘平滑区域的扩散更加精准.经过仿真验证,本文的算法在一定程度上优于Kim算法,可以广泛使用.     

基于混沌映射不变分布的研究

详细地研究了计算混沌映射不变分布的几种数值方法.结果表明,公式法适用于类似抛物线映射简单类的情况;直方图法适用于统计样本较大的情况;特征向量法适用于统计样本较少,但转移概率已知的情况。以logistic map混沌映射为例作了仿真,计算机模拟结果与理论分析结果相符。     

Multivariate outlier detection based on self-organizing map and adaptive nonlinear map and its application

To facilitate visualizing and detecting outliers in high dimensional complex data, a novel method integrating self-organizing map (SOM) with adaptive non-linear map (ANLM) was proposed for multivariate outlier detection. Firstly, the high dimensional complex data are pre-processed by robust scaling. Secondly, SOM is applied to map the pre-processed data onto the SOM plane to obtain the topology of the high dimensional complex data, and then the 2-dimensional projection plane of the trained SOM plane, on which the data distribution can be visualized easily, is obtained via ANLM. In sequel, based on the 2-dimensional plane and the topology, a quasi-3 δ edit rule was proposed to distinguish between the normal data and the outliers in high dimensional complex data. Finally, the proposed multivariate outlier detection was illustrated using synthetic data, two standard benchmark data sets and a real industrial process data. The empirical results show that the outliers in high dimensional complex data are visualized easily on the 2-dimensional plane and effectively detected and eliminated by the quasi-3 δ edit rule, and fully demonstrate its satisfactory ability on dealing with outliers in high dimensional complex data.     

Efficient depth map compression for view rendering in 3D video

Abstract

A depth map represents three-dimensional (3D) scene information, which is used to synthesise virtual views in 3D video coding. Since the quality of synthesised virtual view highly depends on the quality of depth map, efficient depth coding is crucial for the 3D video system. However, depth map is not directly used for display but to generate virtual views. Compressing depth map using existing video coding techniques yields unacceptable distortions in rendered virtual view. Thus, the depth map needs to be compressed in a way that can minimise distortions in the rendered views. To solve this problem, we propose an efficient depth map compression method for view rendering based on view rendering distortion other than depth map distortion itself. Firstly, we derive relationship between distortion in coded depth map and the rendered view. Then, a region-based video characteristic distortion model is proposed to precisely estimate distortion in view synthesis. Finally, a multilateral filtering is applied as an in-loop filter to reduce rendering artefacts. Experimental results show that the proposed method achieves 2 dB coding gain in terms of PSNR and subjective quality improvement of synthesised views.     

基于FPGA的中频采集系统设计

为了实现中频调制信号的分析,根据软件无线电的设计思想,设计实现了基于FPGA的中频采集系统.系统以FPGA器件CycloneⅡEP2C35F67218作为核心处理芯片,利用单片FPGA实现了高速数据流的数据缓存、高速数据存储以及总线数据传输.实验表明,该中频采集系统性能稳定,操作简单,采集数据符合上层软件分析处理的要求,具有广阔的应用前景.     

基于专利地图的企业专利布局设计研究

合理的专利布局设计能够使企业全面掌握行业内的技术实力,准确识别竞争对手,抢占技术制高点.专利地图可以将相关专利信息系统化、直观化、图形化,为了解行业动向、指引研发方向,从而进行专利布局提供依据.从专利布局角度出发并结合专利地图知识,将企业专利布局设计流程简化为4种模型,即专利数据采集模型、企业发展状态模型、竞争对手定位模型和技术潜力发展模型,通过对每一种模型的制作和分析,使企业明确自身的发展状态、竞争者的威胁以及行业技术发展潜力,从而有效地进行专利布局设计,提高竞争力.最后以金刚石工具行业中的某企业为例进行验证.     

Generation of J0-Bessel-Gauss beam by a heterogeneous refractive index map

In this paper, we present the theoretical studies of a refractive index map to implement a Gauss to a J(0)-Bessel-Gauss convertor. We theoretically demonstrate the viability of a device that could be fabricated on a Si/Si(1-y)O(y)/Si(1-x-y)Ge(x)C(y) platform or by photo-refractive media. The proposed device is 200 μm in length and 25 μm in width, and its refractive index varies in controllable steps across the light propagation and transversal directions. The computed conversion efficiency and loss are 90%, and -0.457 dB, respectively. The theoretical results, obtained from the beam conversion efficiency, self-regeneration, and propagation through an opaque obstruction, demonstrate that a two-dimensional (2D) graded index map of the refractive index can be used to transform a Gauss beam into a J(0)-Bessel-Gauss beam. To the best of our knowledge, this is the first demonstration of such beam transformation by means of a 2D index-mapping that is fully integrable in silicon photonics based planar lightwave circuits (PLCs). The concept device is significant for the eventual development of a new array of technologies, such as micro optical tweezers, optical traps, beam reshaping and nonlinear beam diode lasers.     

Generation of vector beams based on dielectric metasurfaces

Vector beam has recently attracted many attentions due to novel properties and wide applications. Finding a method efficiently generating high quality vector beam is very important when it is used in practice. In this paper, we theoretically and experimentally demonstrate the generation of vector beams based on dielectric metasurfaces, which are fabricated by femtosecond laser writing in silica glass. Three types of linearly polarized vector beams are produced by three specially designed metasurfaces. The vector beam generator is convenient and robust due to its simple optical path. We believe that metasurfaces will be widely applied in manipulating polarization, phase, and amplitude of light as the development of fabrication technology.