Detection of moving objects with a moving camera using non-panoramic background model

This paper presents a fast and reliable method for moving object detection with moving cameras (including pan–tilt–zoom and hand-held cameras). Instead of building large panoramic background model as conventional approaches, we construct a small-size background model, whose size is the same as input frame, to decrease computation time and memory storage without loss of detection performance. The small-size background model is built by the proposed single spatio-temporal distributed Gaussian model and this can solve false detection results arising from registration error and background adaptation problem in moving background. More than the proposed background model based on spatial and temporal information, several pre- and post-processing methods are adopted and organized systematically to enhance the detection performances. We evaluate the proposed method with several video sequences under difficult conditions, such as illumination change, large zoom variation, and fast camera movement, and present outperforming detection results of our algorithm with fast computation time.     

Extracting mobile objects by sequential background detection on a video

This paper describes a technique for extracting moving objects from a video image sequence taken by a slowly moving camera as well as a fixed camera. The background subtraction method is effective for extracting moving objects from a video. But the latest background image should be employed for the subtraction in the mobile camera case and in order not to be influenced by the light intensity change. A temporal median technique is proposed in this paper which detects the background at every moment. The camera motion is estimated using a local correlation map and the temporal median filter is applied to the common image area among a set of successive image frames to extract the background. The technique was applied to the video images obtained at a junction from a hand-held camera and those taken at a pedestrians crossing by a camera fixed in a car and successfully detected pedestrians. This work was presented in part at the 13th International Symposium on Artificial Life and Robotics, Oita, Japan, January 31–February 2, 2008     

摄像机运动下基于背景匹配的运动目标检测

提出一种摄像机运动下的运动目标检测技术。首先获取相邻两帧图像上的对应特征点的坐标后应用到仿射变换模型得到全局运动参量来描述背景图像的运动变化。然后根据估计出的全局运动参量通过双线性内插法完成背景匹配。最后采用背景匹配后的两帧图像差进行目标检测。实验证明本算法在提高处理的精确度和时效性的同时还可以处理摄像机发生较大幅度旋转运动情况下的目标检测。     

New trends on moving object detection in video images captured by a moving camera: A survey

This paper presents a survey on the latest methods of moving object detection in video sequences captured by a moving camera. Although many researches and excellent works have reviewed the methods of object detection and background subtraction for a fixed camera, there is no survey which presents a complete review of the existing different methods in the case of moving camera. Most methods in this field can be classified into four categories; modeling based background subtraction, trajectory classification, low rank and sparse matrix decomposition, and object tracking. We discuss in details each category and present the main methods which proposed improvements in the general concept of the techniques. We also present challenges and main concerns in this field as well as performance metrics and some benchmark databases available to evaluate the performance of different moving object detection algorithms.     

Detecting moving objects using the rigidity constraint

A method for visually detecting moving objects from a moving camera using point correspondences in two orthographic views is described. The method applies a simple structure-from-motion analysis and then identifies those points inconsistent with the interpretation of the scene as a single rigid object. It is effective even when the actual motion parameters cannot be recovered. Demonstrations are presented using point correspondences automatically determined from real image sequences     

Detecting moving objects, ghosts, and shadows in video streams

Background subtraction methods are widely exploited for moving object detection in videos in many applications, such as traffic monitoring, human motion capture, and video surveillance. How to correctly and efficiently model and update the background model and how to deal with shadows are two of the most distinguishing and challenging aspects of such approaches. The article proposes a general-purpose method that combines statistical assumptions with the object-level knowledge of moving objects, apparent objects (ghosts), and shadows acquired in the processing of the previous frames. Pixels belonging to moving objects, ghosts, and shadows are processed differently in order to supply an object-based selective update. The proposed approach exploits color information for both background subtraction and shadow detection to improve object segmentation and background update. The approach proves fast, flexible, and precise in terms of both pixel accuracy and reactivity to background changes.     

Land surface identification near Yakutsk in eastern Siberia using video images taken from a hedgehopping aircraft

Aircraft-based remote sensing was carried out over the Lena River region (approximately 100?km horizontal scale) near Yakutsk, in eastern Siberia, on nine days between 24 April and 19 June 2000. A home-use video camera was installed on the aircraft and aerial images of the land surface were recorded. By flying at very low altitudes (100?m and 150?m), we compensated for the low resolution of the camera. In all, 5515 scenes were sampled at 10-s intervals from the video, visually interpreted, and classified. The snow/ice and water cover conditions were also interpreted. The results showed that grassland covers the riverine lowland (RLL) of the Lena River, while the principal form of land cover over the terraces is larch forest (about 61%). There was a remarkable contrast between the left (LBT) and right (RBT) bank terraces of the Lena River, no-forest areas covered 28.8% of the RBT, whereas no-forest areas covered only 13.8% of the LBT. In addition, the LBT had a greater proportion of birch forest than the RBT (15.2% vs 2.5%). The video data over eight days showed that the snow thawed first in the RLL and last in the forested areas of the LBT and RBT. An increase in the area covered by water in the RLL in mid-May was probably due to flooding by the Lena River. Distinguishing insolated from cloud-shadowed scenes of the land surface revealed that the insolation rate was lower over the LBT and RBT than over the RLL. The land-cover database created in this analysis will play an important role as basic, reliable ground-truth information for studies using satellite images.     

Detecting motion regions in presence of strong parallax from a moving camera by multi-view geometric constraints

We present a method for detecting motion regions in video sequences observed by a moving camera in the presence of a strong parallax due to static 3D structures. The proposed method classifies each image pixel into planar background, parallax, or motion regions by sequentially applying 2D planar homographies, the epipolar constraint, and a novel geometric constraint called the "structure consistency constraint." The structure consistency constraint, being the main contribution of this paper, is derived from the relative camera poses in three consecutive frames and is implemented within the "Plane + Parallax" framework. Unlike previous planar-parallax constraints proposed in the literature, the structure consistency constraint does not require the reference plane to be constant across multiple views. It directly measures the inconsistency between the projective structures from the same point under camera motion and reference plane change. The structure consistency constraint is capable of detecting moving objects followed by a moving camera in the same direction, a so-called degenerate configuration where the epipolar constraint fails. We demonstrate the effectiveness and robustness of our method with experimental results of real-world video sequences.     

Passive ranging using a moving camera

This article addresses the problem of determining the 3-dimensional locations of salient points in the environment of a moving camera based on a monocular image sequence obtained by the camera. The camera's translational and rotational velocities are assumed to be known approximately via inertial sensors. The motion of the camera is represented by a constant velocity model. Salient points in the image sequence are extracted using Gabor wavelets and tracked using labeled graph matching. The 3-D positions of the selected environmental points relative to the camera are then estimated recursively using an extended Kalman filter (EKF), after initialization by two-frame motion stereo. The motion parameters of the camera are also refined simultaneously. Experimental results on real data are given. © 1992 John Wiley & Sons, Inc.     

MOVIES: indexing moving objects by shooting index images

With the exponential growth of moving objects data to the Gigabyte range, it has become critical to develop effective techniques for indexing, updating, and querying these massive data sets. To meet the high update rate as well as low query response time requirements of moving object applications, this paper takes a novel approach in moving object indexing. In our approach, we do not require a sophisticated index structure that needs to be adjusted for each incoming update. Rather, we construct conceptually simple short-lived index images that we only keep for a very short period of time (sub-seconds) in main memory. As a consequence, the resulting technique MOVIES supports at the same time high query rates and high update rates, trading this property for query result staleness. Moreover, MOVIES is the first main memory method supporting time-parameterized predictive queries. To support this feature, we present two algorithms: non-predictive MOVIES and predictive MOVIES. We obtain the surprising result that a predictive indexing approach—considered state-of-the-art in an external-memory scenario—does not scale well in a main memory environment. In fact, our results show that MOVIES outperforms state-of-the-art moving object indexes such as a main-memory adapted B ^x -tree by orders of magnitude w.r.t. update rates and query rates. In our experimental evaluation, we index the complete road network of Germany consisting of 40,000,000 road segments and 38,000,000 nodes. We scale our workload up to 100,000,000 moving objects, 58,000,000 updates per second and 10,000 queries per second, a scenario at a scale unmatched by any previous work.     

Depth map calculation for a variable number of moving objects using markov sequential object processes

We advocate the use of Markov sequential object processes for tracking a variable number of moving objects through video frames with a view towards depth calculation. A regression model based on a sequential object process quantifies goodness of fit; regularization terms are incorporated to control within and between frame object interactions. We construct a Markov chain Monte Carlo method for finding the optimal tracks and associated depths and illustrate the approach on a synthetic data set as well as a sport sequence.     

Three-diemensional scanner software using a video camera

This paper presents a software development for the obtaining in an automatic way of three-dimensional models of objects (e.g. sculptures, mechanical pieces). Nowadays, there is available software which is able to create three-dimensional models of objects using laser systems, but they are pretty expensive (up to 120,000 €) due to the high technology used for data capture (laser scanner). On the other hand, the software depicted in this paper develops a new process that captures the data with a simple digital video camera and a motorized pedestal (about 1200 €). This way, more than 1300 images of the object are obtained. The later analysis of the edges of all these images will completely define the geometry of the object. So, this software is able to create accurate digital models with a resolution that varies from hundreds to one million of points. In order to export these three-dimensional models the software generates VRML, or ASCII files. These models are very useful in sciences like computer graphics, architecture, mechanics, automated mapping/facilities management, etc.     

Toward a model-based Bayesian theory for estimating and recognizingparameterized 3-D objects using two or more images taken from differentpositions

A parametric modeling and statistical estimation approach is proposed and simulation data are shown for estimating 3-D object surfaces from images taken by calibrated cameras in two positions. The parameter estimation suggested is gradient descent, though other search strategies are also possible. Processing image data in blocks (windows) is central to the approach. After objects are modeled as patches of spheres, cylinders, planes and general quadrics-primitive objects, the estimation proceeds by searching in parameter space to simultaneously determine and use the appropriate pair of image regions, one from each image, and to use these for estimating a 3-D surface patch. The expression for the joint likelihood of the two images is derived and it is shown that the algorithm is a maximum-likelihood parameter estimator. A concept arising in the maximum likelihood estimation of 3-D surfaces is modeled and estimated. Cramer-Rao lower bounds are derived for the covariance matrices for the errors in estimating the a priori unknown object surface shape parameters     

A robust framework for joint background/foreground segmentation of complex video scenes filmed with freely moving camera

This paper explores a robust region-based general framework for discriminating between background and foreground objects within a complex video sequence. The proposed framework works under difficult conditions such as dynamic background and nominally moving camera. The originality of this work lies essentially in our use of the semantic information provided by the regions while simultaneously identifying novel objects (foreground) and non-novel ones (background). The information of background regions is exploited to make moving objects detection more efficient, and vice-versa. In fact, an initial panoramic background is modeled using region-based mosaicing in order to be sufficiently robust to noise from lighting effects and shadowing by foreground objects. After the elimination of the camera movement using motion compensation, the resulting panoramic image should essentially contain the background and the ghost-like traces of the moving objects. Then, while comparing the panoramic image of the background with the individual frames, a simple median-based background subtraction permits a rough identification of foreground objects. Joint background-foreground validation, based on region segmentation, is then used for a further examination of individual foreground pixels intended to eliminate false positives and to localize shadow effects. Thus, we first obtain a foreground mask from a slow-adapting algorithm, and then validate foreground pixels (moving visual objects + shadows) by a simple moving object model built by using both background and foreground regions. The tests realized on various well-known challenging real videos (across a variety of domains) show clearly the robustness of the suggested solution. This solution, which is relatively computationally inexpensive, can be used under difficult conditions such as dynamic background, nominally moving camera and shadows. In addition to the visual evaluation, spatial-based evaluation statistics, given hand-labeled ground truth, has been used as a performance measure of moving visual objects detection.     

Detecting objects in images in real-time computer vision systems using structured geometric models

Methods for designing high-level image processing algorithms for real-time computer vision systems (VS) in which images of objects or their parts can be represented using structured geometric models that have step-like edge features as their primitive elements.     

一种基于单CCD与经纬仪的树木胸径测量方法

提出一种基于单CCD相机与经纬仪的树木图像的胸径测量方法,利用单CCD相机采集的树木图像信息,经过阈值分割、开运算、边缘提取等步骤提取树木胸径二维坐标;利用经纬仪采集的角度信息转换经纬仪旋转坐标系、相机光心坐标系和图像坐标系,得出胸径的三维坐标,最终计算出胸径尺寸。采集了八组树木胸径,平均绝对误差为736 mm,平均相对误差率为246%,完全满足精度要求,验证了将单CCD相机与经纬仪应用于树木胸径测量的方法是可行的。     

一种自适应的运动目标阴影消除新算法

精确地消除活动阴影对运动目标的影响是智能视频监控的核心任务之一,对此提出了一种基于局部纹理分析的自适应阴影消除新算法。进行了基于高斯混合模型的背景重建,并根据阴影的光学特性进行了阴影区域的预检测,得到疑似阴影区域;提出了一种新的自适应动态纹理分析方法并在此基础上实现了活动阴影的检测与消除。实验结果验证了算法的有效性和实用性。     

Detection of moving foreground objects in videos with strong camera motion

In this paper, we propose a novel method for moving foreground object extraction in sequences taken by a wearable camera, with strong motion. We use camera motion compensated frame differencing, enhanced with a novel kernel-based estimation of the probability density function of background pixels. The probability density functions are used for filtering false foreground pixels on the motion compensated difference frame. The estimation is based on a limited number of measurements; therefore, we introduce a special, spatio-temporal sample point selection and an adaptive thresholding method to deal with this challenge. Foreground objects are built with the DBSCAN algorithm from detected foreground pixels.     

背景重建的运动目标分割改进算法

提出了基于背景重建的运动目标分割改进算法。首先使用多帧差获得初次的背景和前景分离,使用前后多帧差重合部分的合并作为当前帧的运动对象,并通过形态膨胀操作消除二值模板中的空洞,得到该帧的并不精确的初次运动目标掩膜。通过掩膜获得稳定的累积背景。通过得到的背景差和帧间差相结合制定判断原则,获得精确的运动目标掩膜,从而对运动目标进行分割。实验表明该算法对于较大面积的运动目标,或者当摄像机拍摄方向与目标运动方向一致的情形下都可以得到较准确的效果。