Real-time moving object detection for video monitoring systems

Moving object detection is one of the challenging problems in video monitoring systems, especially when the illumination changes and shadow exists. Amethod for real-time moving object detection is described. Anew background model is proposed to handle the illumination varition problem. With optical flow technology and background subtraction, a moving object is extracted quickly and accurately. An effective shadow elimination algorithm based on color features is used to refine the moving obj ects. Experimental results demonstrate that the proposed method can update the background exactly and quickly along with the varition of illumination, and the shadow can be eliminated effectively. The proposed algorithm is a real-time one which the foundation for further object recognition and understanding of video mum＇toting systems.

Robust background subtraction in traffic video sequence

For intelligent transportation surveillance, a novel background model based on Mart wavelet kernel and a background subtraction technique based on binary discrete wavelet transforms were introduced. The background model kept a sample of intensity values for each pixel in the image and used this sample to estimate the probability density function of the pixel intensity. The density function was estimated using a new Marr wavelet kernel density estimation technique. Since this approach was quite general, the model could approximate any distribution for the pixel intensity without any assumptions about the underlying distribution shape. The background and current frame were transformed in the binary discrete wavelet domain, and background subtraction was performed in each sub-band. After obtaining the foreground, shadow was eliminated by an edge detection method. Experimental results show that the proposed method produces good results with much lower computational complexity and effectively extracts the moving objects with accuracy ratio higher than 90%, indicating that the proposed method is an effective algorithm for intelligent transportation system.

Adaptive foreground and shadow segmentation using hidden conditional random fields

Video object segmentation is important for video surveillance, object tracking, video object recognition and video editing. An adaptive video segmentation algorithm based on hidden conditional random fields (HCRFs) is proposed, which models spatio-temporal constraints of video sequence. In order to improve the segmentation quality, the weights of spatio-temporal con- straints are adaptively updated by on-line learning for HCRFs. Shadows are the factors affecting segmentation quality. To separate foreground objects from the shadows they cast, linear transform for Gaussian distribution of the background is adopted to model the shadow. The experimental results demonstrated that the error ratio of our algorithm is reduced by 23% and 19% respectively, compared with the Gaussian mixture model (GMM) and spatio-temporal Markov random fields (MRFs).

A shadow- eliminated vegetation index (SEVI) for removal of self and cast shadow effects on vegetation in rugged terrains

The effect of terrain shadow, including the self and cast shadows, is one ofthe main obstacles for accurate retrieval of vegetation parameters byremote sensing in rugged terrains. A shadow- eliminated vegetation index(SEVI) was developed, which was computed from only red and nearinfrared top-of-atmosphere reflectance without other heterogeneous dataand topographic correction. After introduction of the conceptual modeland feature analysis of conventional wavebands, the SEVI was constructedby ratio vegetation index (RVI), shadow vegetation index (SVI) andadjustment factor (f (Δ)). Then three methods were used to validate theSEVI accuracy in elimination of terrain shadow effects, including relativeerror analysis, correlation analysis between the cosine of solar incidenceangle (cosi) and vegetation indices, and comparison analysis between SEVIand conventional vegetation indices with topographic correction. Thevalidation results based on 532 samples showed that the SEVI relativeerrors for self and cast shadows were 4.32% and 1.51% respectively. Thecoefficient of determination between cosi and SEVI was only 0.032 and thecoefficient of variation (std/mean) for SEVI was 12.59%. The results indicatethat the proposed SEVI effectively eliminated the effect of terrain shadowsand achieved similar or better results than conventional vegetation indiceswith topographic correction.

Unsupervised motion detection with background update and shadow suppression

An algorithm is developed to detect moving object and suppress shadow. According to motion variations caused by some moving objects in a scene, a background update approach is proposed. The developed update method ef- ficiently prevents undesired corruption of background and does not consider the adaptation coefficient or the learning rate used in some existing algorithms. A multi-scale wavelet transform methodology is used to segment foreground from a clut- ter background. The optimal selection of threshold value is automatically determined which does not require any complex supervised training or manual calibration. According to photometric invariant, a color ratio difference is proposed to sup- press shadow. Some complete foreground motion object regions are extracted by integrating moving object segmentation in the multi-scale wavelet with shadow suppression in the color ratio difference. The mentioned method is less affected by the presence of moving objects in a scene. Experimental results show that the proposed approach is efficient in detecting motion objects and suppressing shadows by comparisons.

Framework for Freeway Auto-surveillance from Traffic Video

Video based surveillance systems have been widely used on freeway for traffic monitoring, as the cameras can provide the most intuitionistic information. In order to manage all the traffic videos automatically, in this paper, a distributed real-time auto- surveillance system is presented. The freeway traffic videos are taken as input video from Pan Tilt Zoom （PTZ） camera, and then produces an analysis of the states and activity of the vehicles in the region of interested （ROI）, if there is any abnormal instance, an alarm and corresponding traffic video are sent to awake surveillants by Ethernet. To achieve this functionality, our system relies on three main procedures. The first one initializes the system. It detects the ROI of the scene, and performs the camera calibration to remove the perspective effect of the incoming image. The second one segments moving vehicles from the images, eliminate shadow and tracks them real-time. It uses a set of methods to obtain the background of the image, extracts the moving regions and tracks these moving regions by matching them between frames of the video sequence to obtain high-level information such as color, size, velocity, and trajectories of moving vehicles. In the third procedure, activities of vehicles are analyzed based on a series of preset situations which would happen on freeway. The detail information of each vehicle and the global statistical information are checked to find out any abnormal instance, and then triggered an alarm. We present details of the system, together with experiment results which demonstrate the accuracy and time responses.