一种实时的复杂户外场景下目标检测算法

根据实际系统中高帧频摄像机背景图像运动不大而目标有一定运动的特点,首先分离奇偶帧图像,隔帧相减并用自适应阈值技术二值化图像来检测运动目标,经过形态学处理得到许多连通的区域,然后利用目标面积最大等特征识别目标并用外接矩形定位目标,得到目标运动的真实轨迹。仿真结果表明：该算法能简单有效地检测机动目标且易于硬件并行实现。     

天文图像序列中弱目标的实时检测算法

针对天文图像中运动弱小目标的检测问题,在分析天文CCD图像特点的基础上,根据待检测目标运动状态的不同,提出:1)在检测动目标时,对基于图像对称差分运算方法进行了改进,改进后的方法性能优于图像差分法,且硬件实现容易。该方法以连续三帧序列图像为一组处理对象,在进行绝对差运算之前,对图像进行对比度增强及均值滤波;2)使用形态学滤波的方法实现单帧静止多目标的检测,该方法采用top-hat算子完成背景的估计与目标的检测。为了实时实现所提出的动目标及静止目标的检测算法,设计了DSP FPGA硬件架构方案,并进行了外场实验。实验的结果表明,检测算法在硬件加速的情况下可以实时有效地检测到SNR≈2的弱小目标,并可以同时实时保存原始图像数据。     

摄像机旋转运动下的快速目标检测算法

论文提出了一种摄像机旋转运动下的快速目标检测算法。首先为图像的全局运动建立旋转参数模型,然后基于运动预测在相邻帧之间建立SIFT特征点对,利用RANSAC去除外点的影响,结合最小二乘法求解全局运动参数进行运动补偿,基于残差图像的更新策略实时更新特征点集,以适应背景的变化,最后使用帧差法获得运动目标。该算法不仅保持了SIFT本身的优越性能,而且极大地提高了检测速度。实验结果表明该算法可以实时准确的检测出运动目标。     

复杂背景下运动目标的提取

提出了在复杂背景以及摄像机运动的情况下运动目标提取的算法。采用了新的局部投影的方法,在图像中合理地划分了投影区域,在求投影偏移量时提出了峰值匹配的算法,合理地检测到了投影偏移量。求得各个局部运动矢量后,采用出现频率最高的局部运动矢量代表全局运动矢量,成功的补偿了背景的运动。实验结果表明该算法在图像背景复杂且由于摄像机运动使背号发生明显变化的情况下,可稳定准确地提取出运动目标,并能有效克服背景中存在的运动干扰。     

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.     

随机图像选取与自适应背景更新的运动物体检测

为克服运动目标检测中易出现的光照变化、遮挡、虚假目标等现象,提出了一种随机图像选取与自适应背景更新的运动物体检测方法.该方法从视频序列中随机选取一帧图像作为初始背景,根据变化标记矩阵对背景进行自适应迭代更新,以提取可靠的背景图像,实现运动物体的检测.实验结果表明,采用该算法提取的背景不存在混合现象,且在光照变化较大以及运动物体之间存在遮挡的情况下,能够构造出可靠的背景,检测出的目标物体清晰可见.     

一种新的运动目标检测与跟踪算法

常用的运动目标检测算法无法解决在摄像机运动-目标运动情况下的运动目标检测，为此提出了背景匹配法。通过相关匹配算法使背景对齐，结合帧间差分技术有效地将运动目标提取出来；然后利用卡尔曼预测器对运动目标在图像中的位置进行预测，结合增量式带死区的PID控制算法，控制摄像机对准目标。实验结果表明，匹配块的选择加快了处理速度和提高了算法的稳定性，卡尔曼预测器使得跟踪更为平稳可靠。本方法具有简单、通用、抗噪等特点。     

天基平台深空运动小目标检测

针对天基平台探测深空运动目标的特点,本文简要介绍了天基空间目标监视系统工作在凝视成像模式下检测深空运动小目标的方法,在此基础上探讨了基于图像配准技术检测运动小目标方法在天基空间目标监视系统中的应用.本文提出一种星点间部最小距离特征的概念,构建局部最小距离特征的算法.该方法根据点群间的相对位置关系,采用递归搜索的方法构建特征.实验结果表明,局部最小距离特征在星图中分布均匀,实现了星图序列的快速配准,为基于图像配准技术实现运动目标检测的方法提供了保障.     

基于光电成像的单站被动测距

研究空间运动目标到机动摄像机的距离的估计算法.通过相邻帧图像中同一目标上四对特征点的匹配,计算相邻采样时刻目标相对摄像机的距离变化量,结合目标的方位角、俯仰角、特征线度和摄像机的空间坐标,求解一个四阶非线性方程,得到前一时刻目标与摄像机之间的距离,随即确定出当前时刻上目标到摄像机距离.阐述了测距原理,推导了测距方程,最后给出了相应的实验数据.研究表明该方案通过两幅图像即可实现对目标的被动测距,缩比模型试验证明了算法的正确性和有效性.该方法适用于能够面成像的刚体三维运动目标.     

序列图像中目标的分割与跟踪

本文主要研究序列图像中目标自适应跟踪方法.文章首先分析比较了目前国内外常用的分割方法的优缺点,然后提出了一种可以实现快速跟踪的算法.该方法利用序列连续图像的帧间相关性和差异性检测目标,把目标历史运动信息经过预测滤波后的结果和差值的序列图像信息进行比较,以达到快速跟踪的目的.     

基于时空坐标图像的运动点目标检测方法

针对复杂背景下运动点目标检测方法存在的局限性,提出从时间空间角度检测运动目标的方法.在时间空间域里,背景的复杂纹理表现为无限长直线,运动目标则表现为有限长线段.针对该特点,运用逐点灰度均值将背景与运动点目标分离,并根据投影关系将目标投影到二维空间,获得直观的目标运动轨迹.试验结果表明,该方法能较好地从复杂背景中分离出速度不大于1像素/帧的点目标,且可同时进行多目标的检测,抗噪声能力强.     

Nonlinear correlation for estimating the motion of multiple objects in image sequences

A nonlinear correlation algorithm is proposed for estimating the motion of objects from an image pair. This algorithm requires no a priori information on the number, size, or shape of the moving objects and does not require feature extraction or segmentation of either image. The algorithm directly yields information on the number of moving objects, the motion of the objects, and the size of the objects. Additional processing can be performed to yield the centroid of the objects in either frame. The utility of the resulting algorithm is demonstrated by application to a pair of example image sequences.     

中值滤波的视频背景更新

视频流的一个小时间段的图像序列的中值滤波,可以在动态目标不是特别密集的情况下获得较好的实时更新的背景图像.本文重点针对视频动态目标检测中的背景实时更新问题,采用中值滤波方法,进行了比较深入的研究,并就如何选取图像序列进行中值滤波进行了详细的比较计算,建立了中值滤波的理论模型.研究表明,如果图像序列时间段过短,背景中就会有比较大的目标阴影,如果图像序列时间段过长,则不能反映实时背景,在目标检测中会有较大误差.根据理论模型选取时间段适当的几帧图像进行中值滤波,就可以较好地兼顾背景实时性提取和消除目标阴影的目的.     

Multi-resolution fuzzy clustering approach for image-based particle characterization for particle systems

Online characterization of particles is an important step for maintaining desired product quality in particulate processes. Direct real-time image analysis is a promising method for monitoring particle systems, and is becoming increasingly more attractive due to availability of high speed imaging devices and equally powerful computers. Performing image segmentation (separation of objects (particles) within one image) accurately becomes a key issue in particle image analysis. This paper presents a novel technique based on combining wavelet transform and Fuzzy C-means Clustering (FCM) for particle image segmentation. Through performing wavelet transform on images, the noise and high frequency components of images can be eliminated and the textures and features can be obtained. FCM is then used to divide data into two clusters to separate touching objects. To quantitatively evaluate this method, a case study involving a particle image is investigated. The procedure of selecting optimum wavelet function and decomposition level for this image is presented. ‘Fuzzy range’ is used as a derived feature for segmentation. The number of particles, particle equivalent diameters, and size distribution before and after partition are discussed. The results show that this method is effective and reliable.     

Effects of image restoration on automatic acquisition of moving objects in thermal video sequences degraded by the atmosphere

We aim to determine the effect of image restoration (deblurring) on the ability to acquire moving objects detected automatically from long-distance thermal video signals. This is done by first restoring the videos using a blind-deconvolution method developed recently, and then examining its effect on the geometrical features of automatically detected moving objects. Results show that for modern (low-noise and high-resolution) thermal imaging devices, the geometrical features obtained from the restored videos better resemble the true properties of the objects. These results correspond to a previous study, which demonstrated that image restoration can significantly improve the ability of human observers to acquire moving objects from long-range thermal videos.     

Image retargeting approach for ranging cameras

This paper presents a novel image retargeting approach for ranging cameras. The proposed approach first extracts three feature maps: depth map, saliency map and gradient map. Then, the depth map and the saliency map are used to separate the main contents and the background and thus compute a map of saliency objects. After that, the proposed approach constructs an importance map which combines the four feature maps by the weighted sum. Finally, the proposed approach constructs the target image using the seam carving method based on the importance map. Unlike previous approaches, the proposed approach preserves the salient object well and maintains the gradients and visual effects in the background. Moreover, it protects the salient object from being destroyed by the seam carving algorithm. The experimental results show that the proposed approach performs well in terms of the resized quality.     

车辆号牌识别中车辆速度的估计

光照不足导致了运动物体图像出现运动模糊。为提高图像质量,对多种算法方案进行了分 析,提出了一种稳健实用计算量少的新方法。在车辆号牌识别的实际例子里利用图像中运动物体的特征点估算出运动物体的运动速度从而对图像进行修正处理,成功地得到满意的图像。     

一种视频图像序列中运动对象的分割与跟踪算法

本文提出了一种视频图像序列中运动对象的分割与跟踪算法。该算法通过Canny算子检测出差帧图像的边缘信息,并结合当前帧与背景帧的边缘图像,提取出运动对象。在后续帧中通过建立前帧感兴趣运动对象与当前帧中各运动对象的帧间向量来跟踪当前帧中感兴趣的视频对象。实验结果表明,该算法可行,而且由于该算法简单、计算复杂度小,能很好地满足实时监控系统中对感兴趣运动对象的提取与跟踪。     

基于小波分析的可见光图像自动配准方法研究

研究了一种快速,准确,抗噪的可见光图像自动配方法。首先用小波分析技术提取两幅图像的特征点,然后对两幅图像之间的角度差进行补偿,最后用多层特征点匹配技术完成两幅图像的变换参数的估计,对一定研究领域的可见光图像自动配准的仿真实验表明;该方法可以比较快速,准确,自动地得到这些图像之间的配参数。且对噪声具有一定的适应能力。     

Moving/motionless foreground object detection using fast statistical background updating

Abstract

In video surveillance, the detection of foreground objects in an image sequence from a still camera is very important for object tracking, activity recognition and behaviour understanding. The conventional background subtraction cannot respond promptly to dynamic changes in the background, and temporal difference cannot accurately extract the object shapes and detect motionless objects. In this paper, we propose a fast statistical process control scheme for foreground segmentation. The proposed method can promptly calculate the exact grey-level mean and standard deviation of individual pixels in both short- and long-term image sequences by simply deleting the earliest one among the set of images and adding the current image scene in the image sequence. A short-term updating process can be highly responsive to dynamic changes of the environment, and a long-term updating process can well extract the shape of a moving object. The detection results from both the short- and long-term processes are incorporated to detect motionless objects and eliminate non-stationary background objects. Experimental results have shown that the proposed scheme can be well applied to both indoor and outdoor environments. It can effectively extract foreground objects with various moving speeds or without motion at a high process frame rate.