基于SIFT特征点的无人机影像拼接方法研究

针对无人机空中姿态不稳定,其拍摄影像存在倾斜、曝光不均匀等缺点,采用常规的影像匹配方法效果不佳,甚至无法匹配,而SIFT算法对图像几何变形,分辨率差异、旋转等有较好适应性.利用SIFT算法对无人机影像进行特征点的提取,并采用最小二乘算法对影像进行精匹配.同时,为了减少拼接过程中误差的传播,提出了分块拼接方法.实验结果表...     

基于直方图统计特征的虹膜粗分类

为改善大规模虹膜数据库中虹膜匹配算法的性能,本文提出了一种基于灰度统计特征的虹膜类型识别算法.该算法首先将虹膜纹理区域划分为8个子块,然后利用直方图比率提取虹膜类型的直方图统计特征.最后,根据对应子块特征的相似度将虹膜图像识别为五种类型.所提算法在CASIA虹膜库中挑选的500幅虹膜图像样本上进行了测试.仿真实验结果表...     

一个提高基于网络入侵检测系统规则匹配速度算法的研究

在分析现有基于网络入侵检测系统的规则匹配算法的基础上,提出一种基于哈希函数的规则匹配方法.通过与现有网络入侵检测系统Snort[1]进行比较,证明该系统匹配速度要高于Snort.     

基于互信息特征选择和LSSVM的网络入侵检测系统

为提高网络入侵检测系统(IDS)的性能,提出一种基于互信息特征选择和LSSVM的入侵检测方案(BMIFSLSSVM)。将采集到的网络连接数据进行规范化处理,并提出一种权衡考虑特征相关性和冗余性的新型互信息特征选择(BMIFS)方法,以此从网络连接数据中选择出有效特征集。根据提取的训练样本特征集,利用最小二乘支持向量机(LSSVM)构建分类器和简化粒子群优化(SPSO)算法来优化LSSVM的核函数宽度系数和正则化参数,最后利用训练好的分类器进行入侵检测。仿真结果表明:提出的BMIFS能够选择出最优特征集,使BMIFS-LSSVM提高入侵检测准确率且降低误报率。     

基于WMRMR的滚动轴承混合域特征选择方法

为充分利用时域、频域以及时频域中的有效特征,提高滚动轴承故障诊断准确率,提出一种混合域特征集构建方法,利用原始信号分别生成时域和频域特征集,通过经验模式分解提取固有模态函数的排列熵和Hilbert谱的奇异值作为时频域特征集,使得混合域特征集比单域特征更能全面准确反映轴承运行状态。针对混合域特征集存在维数过高、特征之间冗余性严重的问题,采用加权最大相关最小冗余的特征选择方法,以支持向量机分类正确率为依据,选取7个有效特征向量。实验结果表明:基于WMRMR的混合域特征选择方法的分类准确率可达98%,能够有效的识别轴承故障信息。     

基于手背静脉虹膜和指纹融合身份识别算法

针对单模态生物特征识别的局限性,提出融合手背静脉、虹膜和指纹三种生物特征实现身份识别.首先分别对手背静脉图像、虹膜图像和指纹图像进行独立的图像预处理,特征提取和特征匹配,输出各自的匹配分数.分析匹配分数归一化对识别性能的影响,采用Tarh归一化方法对三种生物特征的匹配分数进行归一化处理,最后利用加权求和法则实现匹配分数的融合,利用最小距离分类器实现身份识别.实验结果表明,融合识别算法的等错率为0.009%,当错误接受率接近0时,对应的错误拒绝率仅为0.2%.     

基于特征点的LUT色彩匹配算法

色彩匹配算法是指对于两张图片（源图像和目标图像）,将源图像中的色彩特性和感觉传递到目标图像上,进而使目标图像和源图像有相近色彩感觉的一种算法。本文提出的色彩匹配算法,可以不由用户进行干预,自动寻找图像中的特征点进行匹配,生成3DI。UT对图像进行色彩匹配。与传统的色彩匹配算法不同,基于特征点的LUT色彩匹配算法在对于立体图对的色彩匹配上,有很好的效果。由于匹配的结果是通过3DLUT来表达的,因此本算法相对于传统的色彩匹配算法,更加适用于影视后期制作环节。     

大规模复杂规则匹配技术研究

针对当前安全系统对复杂规则的需求和复杂规则匹配技术的状况,提出了一种新的规则表示方式——字符串表达式,并给出了对应的匹配方法——基于扩展的有限状态自动机(XFA)实现大规模复杂规则匹配的算法。字符串表达式可以描述多个精确字符串之间的逻辑关系与空间位置关系,从而满足安全系统对复杂特征的描述需求。匹配使用二维结构来完成,首先用经典串匹配算法进行字符串的存在性验证,然后将其结果作为输入,驱动以表达式中字符串为"字符"的XFA完成逻辑关系的验证。基于XFA的匹配方法的空间效率和时间效率都接近多模精确串匹配算法。实验结果表明,文中提出的方法既能满足安全系统对关联特征的描述需求,又能提供高效的匹配性能,较好地解决了大规模(万条)的复杂规则匹配问题。     

遗传特征选择方法在水下目标识别中的应用

为了改善分类器的性能,提高水下目标识别的正确率,文中研究了遗传特征选择方法在水下目标识别中的应用,设计了一个新的合适度函数作为评价特征子集的准则。对wine数据集和海上实测的舰船辐射噪声数据集的仿真实验之结果表明,该方法可以选择出描述目标的有效特征子集,降低了特征集的维数,从而改善了分类器的性能。     

利用径向和环向分布特征的星图识别方法

提出一种基于径向和环向特征的全天自主星图识别方法。该方法利用具有旋转不变性的径向特征作初始匹配,而以环向特征作后续匹配,并用FOV约束进一步剔除冗余匹配。为了加快匹配搜索的速度,采用查找表的方式构建径向模式库。为使观测星图中尽可能多的星找到其对应匹配星,引入了验证识别环节。仿真实验表明,该方法在较高位置噪声水平下(噪声方差为1pixel)仍能达到97.57%的识别率,比同实验条件下的栅格算法提高3%。与传统的方法相比,该方法具有较快的识别速度(18ms)和较小的存储空间(0.344Mb)。     

改进的动态图模型点模式匹配

为了提高概率图模型点模式匹配的精度,本文提出了改进的动态图模型点模式匹配算法。首先,在动态图模型点模式匹配的相似性度量中应用混合高斯分布,以提高模型利用多特征的能力,使匹配方法对噪声更加稳健。其次,在目标点集中引入了虚拟的哑点并给出了包含哑点的相似性度量。当模板中的点和哑点相匹配的相似性度量更大时模板点将和哑点匹配,以减少由异常点所导致的误配。实验结果表明所提出的匹配方法对噪声和异常点更加稳健,匹配的精度也优于传统方法。     

基于点模式匹配的前视目标定位算法

本文提出了一种基于点模式匹配技术的快速和鲁棒性的前视目标定位算法.该算法将前视目标定位问题转化为三维点集和二维点集的匹配问题.建立点集之间一一对应双向约束下的匹配目标函数,通过最小化该目标函数可以同时得到点集之间的匹配矩阵和变换参数.利用确定性退火算法中的退火温度来控制匹配矩阵的模糊度,增强了算法的鲁棒性,减小了陷入局部极小的可能性.实验结果验证了该算法的有效性和鲁棒性.     

大部件对接中基于单位四元数的iGPS测量位姿比对研究

在大尺寸飞机部件数字化对接过程中,iGPS的应用越来越广泛.本文在研究iGPS测量数据结构与形式的基础上,提出了基于单位四元数的iGPS测量位姿比对算法.与目前常用的一些位姿比对方法相比,该算法具有计算简洁明了、编程实现便捷的优点,尤其适用于大旋转角、多点实时测量的情形.验证实验结果表明,在不同旋转角和平移量的情况下,该算法均具有很高的转换精度,可有效适用于iGPS的测量位姿比对.     

重采样和曲面拟合的图像匹配方法

模式识别中的目标匹配和定位是一个传统问题,但是大多数经典算法其定位精度都是象素级的.图像的亚象元匹配算法,可以突破物理分辨率的限制,把匹配和定位精度从象素级提高到亚象元级,从而满足大规模集成电路制造、摄影测量、工业检测和目标检测等应用对精度的要求.将重采样方法和曲面拟合法有机结合的图像亚象元匹配方法,既有重采样方法精度高的优势,同时通过曲面拟合法加快了计算速度,减少了所需时间.实验结果证明了这种基于重采样和曲面拟合的图像亚象元匹配算法的有效性.     

Pattern matching for variation-source identification in manufacturing processes in the presence of unstructured noise

Variation-source identification has received considerable attention from the manufacturing quality improvement community. One widely used method is based on a pattern matching procedure, which identifies process faults by comparing the fault symptom, which is the principal eigenvector of the covariance matrix of the quality measurement, with fault signatures. The presence of unstructured noise as well as the uncertainty due to sampling will cause the direction of the fault symptom to deviate from the corresponding fault signature. The influences of these two effects on pattern matching procedures have previously been studied separately, by assuming either the absence of unstructured noise or the availability of large samples. This paper developes a robust pattern matching procedure that considers both effects simultaneously. Using a machining process as an illustrative example, the paper demonstrates that previous pattern matching procedures can have a remarkably low identification capability when the assumptions are not strictly satisfied. By contrast, our proposed method is more robust, maintaining a good identification probability, and would be a preferable tool for root-cause identification in manufacturing quality improvement.     

Multi-level trajectory learning for traffic behavior detection and analysis

Motion patterns can be learnt automatically based on object trajectories data extracted by means of video tracking, which is an effective approach for modeling and analyzing traffic behavior. In this paper, a multi-level motion pattern learning approach for traffic behavior analysis is presented, which takes into account the spatial characteristics, direction characteristics, and type characteristics of trajectories. At the spatial level, improved Hausdorff distance measurement is applied to construct a spatial similarity matrix of the trajectories collected, and spectral clustering is used to achieve spatial pattern learning. At the directional level, the start and end points of trajectories are fitted using a Gaussian mixed model to extract the distribution of entry and exit zones. Then, the direction pattern is obtained from the regional centers of the pairwise distribution zones. At the type level, the type pattern is acquired through a K-means clustering algorithm that considers multiple classification features of trajectories. Based on the learned multi-level motion patterns, abnormal behavior detection algorithms are further developed by means of pattern matching. Finally, our approach is tested with several video sequences from real-world traffic scenarios. Some typical traffic behaviors in the test scenarios are successfully recognized and analyzed and examples of abnormal traffic behaviors are also reliably detected.     

A study of indoor positioning systems using iBeacons with different transmission power levels

Abstract

The prevalence of smartphones has increased the demand for and application of location-based services. However, the Global Positioning System, currently the most widely used positioning technology, cannot provide accurate positioning services when obstructed by obstacles. Consequently, this system can only provide outdoor application services such as outdoor navigation and tracking. In 2013, Apple Inc. released iBeacon, a positioning technology based on Bluetooth low energy (BLE). This device transmits Bluetooth signals within a specific range, in which the signals are received by other smartphones to calculate distances for providing indoor positioning-related services. In this study, the iBeacon transmission power level is adjusted to significantly increase Bluetooth signal differences in indoor environments. Therefore, it can reduce received signal strength indicators (RSSI) similarity for some reference points by adjusting the power level. Subsequently, radio frequency signals are filtered using a modified moving average filter to reduce signal variations after reception. Next, pattern matching and the K-Nearest Neighbors (KNNs) algorithm are integrated to facilitate positioning. The integration of the modified moving average filter with the KNNs algorithm increases the positioning accuracy by 23.08% during the online phase. This finding can thus improve location-based services.     

基于折页模式与折页单元组配关系的折页机工作效能分析方法

分析了折页模式的描述格式,根据印张折痕的位置关系及变向特性对折页模式进行了分解,将复杂折页模式分解为了一系列平行折的顺次组合。结合平行折研究了折页模式与折页单元之间的组配关系,归纳了折页模式与折页单元的组配规则。以JDF定义的85种常用折页模式为参照,提出了一种基于折页模式与折页单元组配的折页机工作效能分析方法。     

工程图形模式识别中几个关键算法的研究

对工程图形模式识别中的关键问题——细化、细化后处理、矢量化处理问题作了探讨。提出(1)一种新的快速细化算法,利用串行扫描,通过对称侵蚀的方法获取图形目标骨架;(2)采用线模式匹配方式进行细化后处理,恢复图形特征点;(3)一个将二值点阵图形转换为矢量图形的自动跟踪识别算法。以上三种算法均在IBM PC/286计算机上用C语言实现。运行结果表明,对机械图形及其它工程图形进行处理是十分有效的。     

Skipping Undesired High-Frequency Content to Boost DPI Engine

Deep Packet Inspection (DPI) at the core of many monitoring appliances, such as NIDS, NIPS, plays a major role. DPI is beneficial to content providers and censorship to monitor network traffic. However, the surge of network traffic has put tremendous pressure on the performance of DPI. In fact, the sensitive content being monitored is only a minority of network traffic, that is to say, most is undesired. A close look at the network traffic, we found that it contains many undesired high frequency content (UHC) that are not monitored. As everyone knows, the key to improve DPI performance is to skip as many useless characters as possible. Nevertheless, researchers generally study the algorithm of skipping useless characters through sensitive content, ignoring the high-frequency non-sensitive content. To fill this gap, in this literature, we design a model, named Fast AC Model with Skipping (FAMS), to quickly skip UHC while scanning traffic. The model consists of a standard AC automaton, where the input traffic is scanned byte-by-byte, and an additional sub-model, which includes a mapping set and UHC matching model. The mapping set is a bridge between the state node of AC and UHC matching model, while the latter is to select a matching function from hash and fingerprint functions. Our experiments show promising results that we achieve a throughput gain of 1.3- 2.6 times the original throughput and 1.1-1.3 times Barr’s double path method.