基于大数据分析的移动对象轨迹预测方法

对移动对象的轨迹预测将在移动目标跟踪识别中具有较好的应用价值。移动对象轨迹预测的基础是移动目标运动参量的采集和估计,移动目标的运动参量信息特征规模较大,传统的单分量时间序列分析方法难以实现准确的参量估计和轨迹预测。提出一种基于大数据多传感信息融合跟踪的移动对象轨迹预测算法。首先进行移动目标对象进行轨迹跟踪的控制对象描述和约束参量分析,对轨迹预测的大规模运动参量信息进行信息融合和自正整定性控制,通过大数据分析方法实现对移动对象运动参量的准确估计和检测,由此指导移动对象轨迹的准确预测,提高预测精度。仿真结果表明,采用该算法进行移动对象的运动参量估计和轨迹预测的精度较高,自适应性能较强,稳健性较好,相关的指标性能优于传统方法。     

SSD与时空特征融合的视频目标检测

目的 视频目标检测旨在序列图像中定位运动目标,并为各个目标分配指定的类别标签。视频目标检测存在目标模糊和多目标遮挡等问题,现有的大部分视频目标检测方法是在静态图像目标检测的基础上,通过考虑时空一致性来提高运动目标检测的准确率,但由于运动目标存在遮挡、模糊等现象,目前视频目标检测的鲁棒性不高。为此,本文提出了一种单阶段多框检测（single shot multibox detector,SSD）与时空特征融合的视频目标检测模型。方法 在单阶段目标检测的SSD模型框架下,利用光流网络估计当前帧与近邻帧之间的光流场,结合多个近邻帧的特征对当前帧的特征进行运动补偿,并利用特征金字塔网络提取多尺度特征用于检测不同尺寸的目标,最后通过高低层特征融合增强低层特征的语义信息。结果 实验结果表明,本文模型在ImageNet VID （Imagelvet for video object detetion）数据集上的mAP （mean average precision）为72.0%,相对于TCN （temporal convolutional networks）模型、TPN+LSTM （tubelet proposal network and long short term memory network）模型和SSD+孪生网络模型,分别提高了24.5%、3.6%和2.5%,在不同结构网络模型上的分离实验进一步验证了本文模型的有效性。结论 本文模型利用视频特有的时间相关性和空间相关性,通过时空特征融合提高了视频目标检测的准确率,较好地解决了视频目标检测中目标漏检和误检的问题。     

基于复合动态模型和证据融合架构的移动物体检测与跟踪方法

针对现有方法中移动物体检测与跟踪的准确性精度较低的缺点,提出一种基于多传感器检测分类的移动物体描述和感知方法：建立了一个包含核心对象动态特征和分类描述的复合模型,在此基础上设计了一个基于证据框架的信息感知与融合方法,通过整合动态模型和不确定性特征来实现对移动物体的检测和跟踪。为了验证所提方法的有效性,在一辆安装有雷达、激光雷达和摄像头的演示车上进行了相关实验,在不同驾驶场景下针对行人、卡车和轿车三个移动物体进行了检测与跟踪,实验结果证明所提方法具有非常高的准确性。     

Multi-spectral fusion for surveillance systems

Surveillance systems such as object tracking and abandoned object detection systems typically rely on a single modality of colour video for their input. These systems work well in controlled conditions but often fail when low lighting, shadowing, smoke, dust or unstable backgrounds are present, or when the objects of interest are a similar colour to the background. Thermal images are not affected by lighting changes or shadowing, and are not overtly affected by smoke, dust or unstable backgrounds. However, thermal images lack colour information which makes distinguishing between different people or objects of interest within the same scene difficult.By using modalities from both the visible and thermal infrared spectra, we are able to obtain more information from a scene and overcome the problems associated with using either modality individually. We evaluate four approaches for fusing visual and thermal images for use in a person tracking system (two early fusion methods, one mid fusion and one late fusion method), in order to determine the most appropriate method for fusing multiple modalities. We also evaluate two of these approaches for use in abandoned object detection, and propose an abandoned object detection routine that utilises multiple modalities. To aid in the tracking and fusion of the modalities we propose a modified condensation filter that can dynamically change the particle count and features used according to the needs of the system.We compare tracking and abandoned object detection performance for the proposed fusion schemes and the visual and thermal domains on their own. Testing is conducted using the OTCBVS database to evaluate object tracking, and data captured in-house to evaluate the abandoned object detection. Our results show that significant improvement can be achieved, and that a middle fusion scheme is most effective.     

多信息动态融合的运动目标检测方法

针对基于视觉显著性的运动目标检测算法存在时空信息简单融合及忽略运动信息的问题,提出一种动态融合视觉显著性信息和运动信息的运动目标检测方法。该方法首先计算每个像素的局部显著度和全局显著度,并通过贝叶斯准则生成空间显著图;然后,利用结构随机森林算法预测运动边界,生成运动边界图;其次,根据空间显著图和运动边界图属性的变化,动态确定最佳融合权值;最后,根据动态融合权值计算并标记运动目标。该方法既发挥了显著性算法和运动边界算法的优势,又克服了各自的不足,与传统背景差分法和三帧差分法相比,检出率和误检率的最大优化幅度超过40%。实验结果表明,该方法能够准确、完整地检测出运动目标,提升了对场景的适应性。     

Fast moving object detection with non-stationary background

The detection of moving objects under a free-moving camera is a difficult problem because the camera and object motions are mixed together and the objects are often detected into the separated components. To tackle this problem, we propose a fast moving object detection method using optical flow clustering and Delaunay triangulation as follows. First, we extract the corner feature points using Harris corner detector and compute optical flow vectors at the extracted corner feature points. Second, we cluster the optical flow vectors using K-means clustering method and reject the outlier feature points using Random Sample Consensus algorithm. Third, we classify each cluster into the camera and object motion using its scatteredness of optical flow vectors. Fourth, we compensate the camera motion using the multi-resolution block-based motion propagation method and detect the objects using the background subtraction between the previous frame and the motion compensated current frame. Finally, we merge the separately detected objects using Delaunay triangulation. The experimental results using Carnegie Mellon University database show that the proposed moving object detection method outperforms the existing other methods in terms of detection accuracy and processing time.     

基于Transformer的跨尺度交互学习伪装目标检测

伪装目标检测(COD)旨在精确且高效地检测出与背景高度相似的伪装物体, 其方法可为物种保护、医学病患检测和军事监测等领域提供助力, 具有较高的实用价值. 近年来, 采用深度学习方法进行伪装目标检测成为一个比较新兴的研究方向. 但现有大多数COD算法都是以卷积神经网络(CNN)作为特征提取网络, 并且在结合多层次特征时, 忽略了特征表示和融合方法对检测性能的影响. 针对基于卷积神经网络的伪装目标检测模型对被检测目标的全局特征提取能力较弱问题, 提出一种基于Transformer的跨尺度交互学习伪装目标检测方法. 该模型首先提出了双分支特征融合模块, 将经过迭代注意力的特征进行融合, 更好地融合高低层特征; 其次引入了多尺度全局上下文信息模块, 充分联系上下文信息增强特征; 最后提出了多通道池化模块, 能够聚焦被检测物体的局部信息, 提高伪装目标检测准确率. 在CHAMELEON、CAMO以及COD10K数据集上的实验结果表明, 与当前主流的伪装物体检测算法相比较, 该方法生成的预测图更加清晰, 伪装目标检测模型能取得更高精度.     

Learning Novel Objects for Extended Mobile Manipulation

We propose a method for learning novel objects from audio visual input. The proposed method is based on two techniques: out-of-vocabulary (OOV) word segmentation and foreground object detection in complex environments. A voice conversion technique is also involved in the proposed method so that the robot can pronounce the acquired OOV word intelligibly. We also implemented a robotic system that carries out interactive mobile manipulation tasks, which we call “extended mobile manipulation”, using the proposed method. In order to evaluate the robot as a whole, we conducted a task “Supermarket” adopted from the RoboCup@Home league as a standard task for real-world applications. The results reveal that our integrated system works well in real-world applications.     

Real-time abandoned and stolen object detection based on spatio-temporal features in crowded scenes

Abandoned and stolen object detection is a challenging task due to occlusion, changes in lighting, large perspective distortion, and the similarity in appearance of different people. This paper presents real-time detection methods of abandoned and stolen objects in a complex video. The adaptive background modeling method is applied to stable tracking and the ghost image removing. To detect abandoned and stolen objects, the methods determine spatio-temporal relationship between moving people and suspicious drops. The space first detection method measures the distance between a moving object and a non-moving object in spatial change analysis. The time first detection method conducts temporal change analysis and then spatial change analysis. The potential abandoned object is classified as a definite abandoned or stolen object by two-level detection approach. The time-to-live timer is applied by adjusting several key parameters on each camera and environment. In experiments, we show the experimental results to evaluate our proposed methods using benchmark datasets.     

Moving Cast Shadows Detection Using Ratio Edge

Moving objects segmentation plays a very important role in real-time image analysis. However, as one of the common parts in the natural scenes, shadows severely interfere with the accuracy of moving objects detection in video surveillance. In this paper, we present a novel method for moving cast shadows detection. Based on the analysis of the physical model of moving shadows, we prove that the ratio edge is illumination invariant. The distribution of the ratio edge is discussed and a significance test is performed to classify each moving pixel into foreground object or moving shadow. Intensity constraint and geometric heuristics are imposed to further improve the performance. Experiments on various typical scenes exhibit the robustness of the proposed method. Extensively quantitative evaluation and comparison demonstrate that the proposed method significantly outperforms state-of-the-art methods.     

基于非对称跨模态融合的RGB-D显著目标检测

目前大多数RGB-D显著目标检测方法在RGB特征和Depth特征的融合过程中采用对称结构,对两种特征进行相同的操作,忽视了RGB图像和Depth图像的差异性,易造成错误的检测结果.针对该问题,提出一种基于非对称结构的跨模态融合RGB-D显著目标检测方法,利用全局感知模块提取RGB图像的全局特征,并设计了深度去噪模块滤除低质量Depth图像中的大量噪声;再通过所提出的非对称融合模块,充分利用两种特征间的差异性,使用Depth特征定位显著目标,用于指导RGB特征融合,补足显著目标的细节信息,利用两种特征各自的优势形成互补.通过在4个公开的RGB-D显著目标检测数据集上进行大量实验,验证所提出的方法优于当前的主流方法.     

运动目标的自动空时分割算法

在视频应用中,运动目标的提取是一个重要的研究课题。为了对运动目标进行更有效的分割,提出了一种从视频序列中自动提取运动目标的空时分割算法。该算法在时域分割中采用基于齐异矢量消除的目标检测方法来获得运动目标的初始模板。通常,该初始模板具有不连续的边界和一些"孔"。为了得到较为完整的目标区域,用具有距离约束的区域生长算法来补偿初始模板。而在空域分割中,分水岭分割则通过考虑全局信息来增强其分割的精确性。然后,精确的运动目标即可通过空时融合模块提取出来。试验结果表明,该空时分割算法是有效的。     

基于分块和改进粒子滤波的运动目标检测方法

为了快速准确地检测到视频场景中的运动目标,提出了一种基于分块和改进的粒子滤波的运动目标检测方 法。首先,对视频图像序列分块并提取每个图像块的颜色特征;然后,用改进的粒子滤波对每个图像块进行操作,计算 出每个块对应的粒子的权重;最后,根据粒子的平均权重建立背景模型,提取运动目标。将分块和粒子滤波相结合,能 够在不降低检测精度的基础上,大幅减少算法的计算量,提高算法的执行速度。实验结果表明,该方法具有较好的鲁 棒性、杭噪性和抗光照变化能力,提取的运动目标更加完整。     

基于视觉和文本的多模态文档图像目标检测

由于文档图像的布局复杂、目标对象尺寸分布不均匀,现有的检测算法很少考虑多模态信息和全局依赖关系,提出了基于视觉和文本的多模态文档图像目标检测方法。首先探索多模态特征的融合策略,为利用文本特征,将图像中文本序列信息转换为二维表征,在文本特征和视觉特征初次融合之后,将其输入到骨干网络提取多尺度特征,并在提取过程中多次融入文本特征实现多模态特征的深度融合;为保证小物体和大物体的检测精度,设计了一个金字塔网络,该网络的横向连接将上采样的特征图与自下而上生成的特征图在通道上连接,实现高层语义信息和低层特征信息的传播。在大型公开数据集PubLayNet上的实验结果表明,该方法的检测精度为95.86%,与其他检测方法相比有更高的准确率。该方法不仅实现了多模态特征的深度融合,还丰富了融合的多模态特征信息,具有良好的检测性能。     

采用测地线活动轮廓模型检测与跟踪运动目标

水平集几何活动轮廓模型能较好地适应曲线的拓扑变化.为了跟踪和获取刚体和非刚体运动目标的轮廓信息,提出了一种基于改进测地线活动轮廓(GAC)模型和Kalman滤波相结合的算法以检测和跟踪运动目标.该算法首先采用高斯混合模型和背景差分获取目标的运动区域,在运动区域内采用引入距离规则化项的GAC模型进行曲线演化,使改进GAC模型在运动目标的真实轮廓处收敛;然后通过结合Kalman滤波预测目标下一帧的位置,实现对目标轮廓跟踪.实验结果表明,该方法适用于刚体和非刚体目标,在部分遮挡的情况下也能保持良好的检测和跟踪效果.     

各向异性导向滤波的红外与可见光图像融合

目的 针对红外与可见光图像融合时易产生边缘细节信息丢失、融合结果有光晕伪影等问题，同时为充分获取多源图像的重要特征，将各向异性导向滤波和相位一致性结合，提出一种红外与可见光图像融合算法。方法 首先，采用各向异性导向滤波从源图像获得包含大尺度变化的基础图和包含小尺度细节的系列细节图；其次，利用相位一致性和高斯滤波计算显著图，进而通过对比像素显著性得到初始权重二值图，再利用各向异性导向滤波优化权重图，达到去除噪声和抑制光晕伪影；最后，通过图像重构得到融合结果。结果 从主客观两个方面，将所提方法与卷积神经网络（convolutional neural network，CNN）、双树复小波变换（dual-tree complex wavelet transform，DTCWT）、导向滤波（guided filtering，GFF）和各向异性扩散（anisotropic diffusion，ADF）等4种经典红外与可见光融合方法在TNO公开数据集上进行实验对比。主观分析上，所提算法结果在边缘细节、背景保存和目标完整度等方面均优于其他4种方法；客观分析上，选取互信息（mutual information，MI）、边缘信息保持度（degree of edge information，Q^AB/F）、熵（entropy，EN）和基于梯度的特征互信息（gradient based feature mutual information，FMI_gradient）等4种图像质量评价指数进行综合评价。相较于其他4种方法，本文算法的各项指标均有一定幅度的提高，MI平均值较GFF提高了21.67%，Q^AB/F平均值较CNN提高了20.21%，EN平均值较CNN提高了5.69%，FMI_gradient平均值较GFF提高了3.14%。结论 本文基于各向异性导向滤波融合算法可解决原始导向滤波存在的细节"光晕"问题，有效抑制融合结果中伪影的产生，同时具有尺度感知特性，能更好保留源图像的边缘细节信息和背景信息，提高了融合结果的准确性。     

Coupled multi-object tracking and labeling for vehicle trajectory estimation and matching

Efficient detection and tracking of moving objects in real life conditions is a very challenging research issue, mainly due to occlusions, illumination variations, appearance (disappearance) of new (existing) objects and overlapping issues. In this paper, we address these difficulties by incorporating non-linear and recursive identification mechanisms in motion-based detection and tracking algorithms. Non-linearity allows correct identification of object of complex visual properties while the adaptability makes the proposed scheme able to update its behaviour to the dynamic environmental changes. In addition, in this paper, we introduce the concept of polar spectrum which is a measure for determining the deviation of a vehicle trajectory from an ideal trace. The proposed methods (object tracking and trajectory matching) are applied in survey engineering problems dealing with safe design road turns. In particular, the automatically detected trajectory of a moving vehicle is compared with the ideal trace, through the polar spectrum measure, to determine the safety of a road turn. This trace is also compared with the one manually derived using photogrammetric algorithms and a small error is obtained verifying the efficiency of the method.     

基于Shearlet变换域改进的IR与灰度VIS图像融合算法

针对多尺度变换域内红外(IR)与灰度可见光(VIS)图像融合后图像清晰度差、纹理信息不丰富等问题,提出一种基于剪切波变换(ST)域改进的IR与灰度VIS图像融合算法.首先,采用形态学顶帽变换(MTH)增强IR与VIS图像;然后,对增强后的IR与VIS图像采用ST变换,将其分解成高、低频图像,针对高频图像提出局部标准差(LSTD)与系数绝对值的融合策略;针对低频图像提出一种改进的权值融合策略;最后,通过逆剪切波变换(IST)获得最终融合图像.仿真实验结果表明,所推荐的方法具有优越的性能.     

非凸加权核范数及其在运动目标检测中的应用

目的 近年来,低秩矩阵分解被越来越多的应用到运动目标检测中。但该类方法一般将矩阵秩函数松弛为矩阵核函数优化,导致背景恢复精度不高;并且没有考虑到前景目标的先验知识,即区域连续性。为此提出一种结合非凸加权核范数和前景目标区域连续性的目标检测算法。方法 本文提出的运动目标检测模型以鲁棒主成分分析(RPCA)作为基础,在该基础上采用矩阵非凸核范数取代传统的核范数逼近矩阵低秩约束,并结合了前景目标区域连续性的先验知识。该方法恢复出的低秩矩阵即为背景图像矩阵,而稀疏大噪声矩阵则是前景目标位置矩阵。结果 无论是在仿真数据集还是在真实数据集上,本文方法都能够取得比其他低秩类方法更好的效果。在不同数据集上,该方法相对于RPCA方法,前景目标检测性能提升25%左右,背景恢复误差降低0.5左右;而相对于DECOLOR方法,前景目标检测性能提升约2%左右,背景恢复误差降低0.2左右。结论 矩阵秩函数的非凸松弛能够比凸松弛更准确的表征出低秩特征,从而在运动目标检测应用中更准确的恢复出背景。前景目标的区域连续性先验知识能够有效地过滤掉非目标大噪声产生的影响,使得较运动目标检测的精度得到大幅提高。因此,本文方法在动态纹理背景、光照渐变等较复杂场景中均能够较精确地检测出运动目标区域。但由于区域连续性的要求,本文方法对于小区域多目标的检测效果不甚理想。