基于贝叶斯理论嵌套抽样的结构物理参数识别研究EI北大核心CSCD

基于贝叶斯估计的结构物理参数识别中,传统马尔可夫蒙特卡洛抽样(MCMC)在解决高维密度函数问题时往往存在抽样效率低、不收敛等问题。采用嵌套抽样方法代替传统的马尔可夫蒙特卡洛抽样,解决了结构物理参数识别中高维后验联合概率密度函数问题。首先从结构加速度时程响应时程出发,建立了后验联合概率密度函数,然后重新定义了结构参数先验分布与似然函数,实现了基于嵌套抽样的结构物理参数识别。采用该方法分别对10层剪切结构数值模型与3层RC框架结构振动台试验模型进行识别,得到了结构刚度及阻尼比等参数,并与试验现象进行了对比。结果表明,该方法可以解决贝叶斯公式高维后验联合概率密度函数问题,且能高效识别结构物理参数,同时也验证了该方法在真实结构物理参数识别与结构损伤识别中的适用性与可靠性。     

基于贝叶斯信息融合的测量不确定度评定与实时更新

基于贝叶斯信息融合与统计推断原理,建立不确定度动态评定模型,对测量不确定度进行实时更新。引入最大熵原理和爬山搜索优化算法,确定先验分布概率密度函数及样本信息似然函数,结合贝叶斯公式求出后验分布概率密度函数,实现不确定度的优化估计。仿真及实例分析表明,基于贝叶斯和最大熵方法评定及更新的测量不确定度更加接近理论值。     

基于均值偏移算法的双摄像机目标跟踪

通常基于摄像机的目标跟踪都是在单个静止摄像机的假设下进行的,而现实往往有更复杂的状况,需要多角度以及对目标的动态跟踪,多个摄像机的目标跟踪和交接必然成为新的研究热点。本文采用双PTZ摄像机对目标跟踪,建立了一种目标跟踪的实验原型,在这种原型下成功实现对目标的跟踪和交接。实验中采用了改进的均值偏移算法,对摄像机移动具有一定的鲁棒性。     

基于最大似然的分布式多视角目标跟踪算法

针对多摄像机视频跟踪的应用场景,提出了一种具有分布式计算特点的目标跟踪算法.该算法首先利用最大似然准则对多视角目标跟踪问题进行建模,并对似然函数进行分解,最后借助消息传递机制高效求解出跟踪目标在各个摄像机视野中位置的估计值,而不再需要将图像数据传输到中央处理单元进行集中式处理.同时,通过融合多个视角的图像信息,有效地抑制了遮挡现象给跟踪结果造成的影响.定性和定量实验均证明了该算法的有效性.     

基于非稀疏多核支持向量机的重现行人识别

目前,视频跟踪正向大范围长时间目标跟踪研究方向发展。重现行人识别是对行人目标进行大范围长时间持续跟踪的关键技术,是后续行为分析的基础。本文提出了一种基于非稀疏多核支持向量机的重现行人识别算法。首先,该方法提取跟踪行人视频图像序列的多层SIFT视觉单词树特征和多层颜色直方图特征。接着,利用高效的非稀疏多核支持向量机算法在线融合多层SIFT视觉单词树特征和多层颜色直方图特征得到行人外观模型。最后利用存储的行人外观模型库对重现行人进行识别。该方法可应用于多摄像机视频监控中同一行人目标的跨摄像机跟踪以及单摄像机监控中行人目标重新出现的识别。实验结果表明,该方法能快速训练人体目标外观模型,能获得很高的识别率。     

紊流激励下基于贝叶斯统计推断的颤振边界预测方法研究

提出将贝叶斯统计推断方法推广应用于大气紊流激励下飞行器结构的颤振分析，对含不确定性因素影响的模态参数识别与颤振边界预测进行研究。在采用自然激励技术从结构在大气紊流激励下的响应中提取自由衰减信号后，基于贝叶斯统计推断，通过马尔科夫链蒙特卡罗(Markov chain Monte Carlo, MCMC)算法对结构模态参数的后验概率密度函数进行采样识别，并利用Z-W(Zimmerman-Weissenburger)颤振裕度法获取颤振速度概率分布，预测颤振边界并分析其不确定性。进行了数值仿真研究，对大气紊流激励下的结构响应数据进行分析，验证了所提出方法的有效性。     

无重叠视域中多目标最优路径集合的数据关联

针对无重叠视域中难以将运动目标与时空因素发生关联或关联后难以求解问题,提出了采用最优路径的数据关联算法并用离散蚁群算法进行了求解。算法首先利用贝叶斯网络,将目标外观匹配相似度、空间约束和时间约束三者融合,把数据关联问题转换为网络中最优路径的选择问题;其次,把路径间样本对的平均相似度设为评价函数,评价函数取最大值时的路径就是最优路径;最后,根据目标的出现在时间和空间存在离散性的特点,用离散粒子群算法求解最优路径,并用粒子编码记录目标运动路径。本算法在由五个摄像机构成的网络中对运动目标进行跟踪仿真,结果表明能有效地求解多目标的最优路径集合,获取了目标在网络中的运动轨迹,实现了接力跟踪,具有良好的鲁棒性。     

运动摄像机下快速提取运动目标的新方法

本文研究摄像机和目标同时运动情况下的实时目标提取问题.首先运用背景差方法,检测出静止摄像机下的运动区域,为了克服连通域分析法耗时长的不足,提出重心偏移迭代法快速获得感兴趣运动目标.在改进Camshifi跟踪算法中,提出采用Bayesian概率法则在由Kalman滤波器预测的感兴趣区域(ROI)内获取颜色概率密度分布图像(CPDDI),引入即时背景(IB)以抑制背景特征.提出依据跟踪结果进行目标提取的方法,即结合CPDDI特征,并辅以适当的形态学滤波策略,从跟踪结果中提取出运动摄像机下的运动目标,解决目标被动态背景干扰的问题.实验结果表明,提出的算法能够较稳定和完整地提取出运动摄像机下的运动目标,对复杂动态背景的适应性较强,且算法完全达到了实时的运行速度.     

结构影响线识别：反问题可识别性分析与降维贝叶斯不确定性量化

结构影响线识别是移动荷载下既有结构评估的理论基础,其本质上是基于系统输入-输出含噪数据反向对静力系统指定截面的响应函数进行识别。已有研究虽然取得了进展,但它们在以下两个方面存在局限性：缺乏反问题可识别性分析;缺乏不确定性量化。反问题可识别性分析是为了厘清系统识别的参数的解的情况。不确定性量化是基于测量输入-输出含噪数据估计影响线参数的后验概率密度函数。针对上述两个局限性,该文在贝叶斯概率框架的基础上开展关于影响线识别的反问题可识别性分析与贝叶斯不确定性量化。该文进行基于直接参数化的影响线识别,包括系统输入与输出、反问题可识别性分析、参数最优值。经分析得出：一方面,直接参数化无法保证全局模型可识别;另一方面,现有方法即使是全局模型可识别的情况下也无法进行不确定性量化。为保证反问题是全局模型可识别且同时获取参数后验概率密度函数,该文提出基于降维贝叶斯不确定性量化的影响线后验识别,包括系统输入与输出重构、反问题可识别性分析、后验概率密度函数。该文进行模拟数据下新光大桥吊杆拉力影响线识别,与实测及模拟数据下简支梁桥应变影响线识别,验证提出方法的有效性。     

基于在线学习的自适应粒子滤波视频目标跟踪

传统的粒子滤波视觉跟踪算法采用固定模型和大量粒子表征目标后验概率,不能满足复杂条件下的视频目标实时跟踪.为了提高跟踪的鲁棒性和稳定性及计算效率,本文提出将自适应状态演化方程和在线增量学习观测似然模型嵌入到粒子滤波算法;并采用在线自动调整粒子数目的策略,提高粒子滤波视觉跟踪的计算效率.室内外实验结果表明,文中提出的视觉跟踪算法不仅能准确、高效地跟踪序列图像中的运动目标,而且对光照、姿态变化引起的目标表观变化具有良好的鲁棒性.     

雷达与经纬仪的联合定位方法

在测距雷达与经纬仪组成的测量网络中,为了提高目标的跟踪定位能力和获取更多的弹道数据,提出了雷达与经纬仪的联合定位方法.根据雷达与经纬仪实时跟踪目标时的几何关系,推导出目标定位的计算公式.将经纬仪的测角信息和雷达测距信息进行数据融合,得到目标相对于经纬仪的距离,进而确定目标的三维位置信息.分析了定位误差的主要来源,明确了该方法的适用范围.在雷达与经纬仪跟踪远距离目标过程中加以应用.结果表明,雷达与经纬仪间的测量基线越小,该方法的定位误差越接近雷达的测距误差.     

Multiple Pedestrian Detection and Tracking in Night Vision Surveillance Systems

Pedestrian detection and tracking are vital elements of today’s surveillance systems, which make daily life safe for humans. Thus, human detection and visualization have become essential inventions in the field of computer vision. Hence, developing a surveillance system with multiple object recognition and tracking, especially in low light and night-time, is still challenging. Therefore, we propose a novel system based on machine learning and image processing to provide an efficient surveillance system for pedestrian detection and tracking at night. In particular, we propose a system that tackles a two-fold problem by detecting multiple pedestrians in infrared (IR) images using machine learning and tracking them using particle filters. Moreover, a random forest classifier is adopted for image segmentation to identify pedestrians in an image. The result of detection is investigated by particle filter to solve pedestrian tracking. Through the extensive experiment, our system shows 93% segmentation accuracy using a random forest algorithm that demonstrates high accuracy for background and roof classes. Moreover, the system achieved a detection accuracy of 90% using multiple template matching techniques and 81% accuracy for pedestrian tracking. Furthermore, our system can identify that the detected object is a human. Hence, our system provided the best results compared to the state-of-art systems, which proves the effectiveness of the techniques used for image segmentation, classification, and tracking. The presented method is applicable for human detection/tracking, crowd analysis, and monitoring pedestrians in IR video surveillance.     

基于假设理论的序列图像中多目标检测和跟踪

本文提出一种基于目标检测的多维假设多目标检测和跟踪方法,此算法对于序列图像的照明变化和遮蔽现象具有很高的鲁棒性.首先,对序列图像进行背景抑制、时域滤波和杂波剔除的预处理,得到单帧的初始目标检测结果.对于初始检测中存在的漏检和误检现象,采用基于假设理论的跟踪概率模型优化初始检测轨迹;将目标的跟踪信息反馈于目标检测模块,形成一闭环自适应跟踪系统,达到多目标的最优检测和跟踪.实验结果表明了所提出的方法在多目标检测和跟踪中的可行性和有效性.     

二自由度云台的机器视觉控制

本文建立了基于机器视觉反馈的云台控制模型,并通过相邻帧间的目标位置预测来简化搜索算法,实现快速高效的视觉云台对运动目标的连续跟踪.在制定系统空间转角输入量的控制策略中,简化了控制模型并对产生的偏差量进行了分析.结果表明,简化后的控制算法能够实现快速反馈和像面调整,在25帧/秒时实现对运动目标稳健跟踪,且跟踪精度高于传统算法;在目标速度2 m/s、相机60帧/秒时,改进算法实现了对运动目标的连续跟踪.同时提出了立体视觉系统模型及其在云台控制中的应用,通过像面坐标获取空间位置信息以规避目标跟丢现象.     

遮挡和光照变化下的单鱼目标跟踪

鱼类跟踪是对鱼类行为分析的基础,在对水中的鱼类跟踪时,目标鱼会由于姿势的变化、受到周围鱼或者物体的遮挡或者光照的影响变得难以追踪.针对这种情况,提出一种利用MobileNet-SSD(SSD,single shot multibox detector)与Dlib关联跟踪器相结合对鱼体进行跟踪的方法,通过SSD算法精准地...     

仿生学的气动柔性手指逆运动学算法研究

针对手指逆运动学问题,采用了基于贝叶斯网理论的仿生学研究方法.首先对人类手指进行了运动学建模,并通过多组实验,采样记录人手在抓取时,食指各关节在不同时刻的角度;其次,以采样数据为样本,进行贝叶斯网络学习,获得各关节角度之间的内在关系模型;最后将该模型应用于气动柔性手指控制系统.实验表明,该模型解决了多自由度机器人逆运动学冗余度的同时,使得气动柔性手指的运动更接近人类.     

A multimodal approach for individual tracking of people and their belongings

Abstract

In this study, a fully automatic surveillance system for indoor environments which is capable of tracking multiple objects using both visible and thermal band images is proposed. These two modalities are fused to track people and the objects they carry separately using their heat signatures and the owners of the belongings are determined. Fusion of complementary information from different modalities (for example, thermal images are not affected by shadows and there is no thermal reflection or halo effect in visible images) is shown to result in better object detection performance. We use adaptive background modeling and local intensity operation for object detection and the mean-shift tracking algorithm for fully automatic tracking. Trackers are refreshed to resolve potential problems which may occur due to the changes in object’s size, shape and to handle occlusion-split and to detect newly emerging objects as well as objects that leave the scene. The proposed scheme is applied to the abandoned object detection problem and the results are compared with the state of art methods. The results show that the proposed method facilitate individual tracking of objects for various applications, and provide lower false alarm rates compared to the state of art methods when applied to the abandoned object detection problem.     

Uncertainty assessment for reconstructions based on deformable geometry

Deformable geometric models can be used in the context of Bayesian analysis to solve ill-posed tomographic reconstruction problems. The uncertainties associated with a Bayesian analysis may be assessed by generating a set of random samples from the posterior, which may be accomplished using a Markov Chain Monte Carlo (MCMC) technique. We demonstrate the combination of these techniques for a reconstruction of a two-dimensional object from two orthogonal noisy projections. The reconstructed object is modeled in terms of a deformable geometrically defined boundary with a uniform interior density yielding a nonlinear reconstruction problem. We show how an MCMC sequence can be used to estimate uncertainties in the location of the edge of the reconstructed object. © 1997 John Wiley & Sons, Inc. Int J Imaging Syst Technol, 8, 506–512, 1997     

All-optical subtracted joint transform correlator with a holographic interferometer

A new, to our knowledge, subtracted joint transform correlator (SJTC) is proposed that has no digital processing in a computer. All processing for obtaining correlation signals between an object and multiple reference patterns were treated optically by use of a joint transform correlator with a holographic interferometer similar to the Mach-Zehnder one. The joint power spectrum of the reference patterns was subtracted from that of the input image (the object pattern plus the reference patterns), and the spurious correlation signals between the different reference patterns were removed. Because of the optical parallel computations of the Fourier spectra and the subtraction, the real-time SJTC is possible to achieve by use of only an optical system. An experimental arrangement of the system and system performances of the shift-invariant characteristics and discriminability are described. The results show the good performance of this system.     

Bionic tracking method by hand & eye-vergence visual servoing

Animals rotate their eyes to gaze at the target prey, enhancing the ability of measuring the distance to the target precisely for catching it. These animals, visual tracking includes the triangular eye-vergence control and their body's motion control by visual servoing. The research aims to realize a bionic robot tracking performance, in which the body links moves together with eyes' view orientation. This paper proposed a hand & eye-vergence dual control system which included two feedback loops: an outer loop for conventional visual servoing to direct a manipulator toward a target object and an inner loop for active motion control of binocular cameras to change the viewpoint along with the moving object to give an accurate and broad observation. This research also foused on how to compensate a fictional motion of the target seen by camera images in an eye-in-hand system, where the camera was fixed on the end-effector and moved together with the hand motion. A robust motion-feedforward (MFF) recognition method is proposed to compensate the fictional motion of the target based on the manipulator's joint velocity, then the real motion of the target seen by camera images is extracted, which can improve the feedback image sensing unit to make the whole servoing system dynamically stable. The effectiveness of the proposed hand & eye-vergence visual servoing method is shown by tracking experiments using a 6-DoF robot manipulator and a 3-DoF binocular vision system.