Motion Detection Based on Background Modeling and Performance Analysis for Outdoor Surveillance

The analysis of moving objects in video sequences has been a paramount issue in applications related to intelligent surveillance systems, robotics, and medicine. Although several works aimed to analyze objects in video sequences have been reported, many of them need manual parameter adjustments and they are not tolerant to illumination changes and dynamic backgrounds. Therefore, a novel scheme termed Dynamic Retinotopic SOM based on an adaptive artificial neural network, to detect moving objects is proposed in this work. The neural network is a model based on the mechanisms of the visual cortex that we called Retinotopic SOM (RESOM) and it is also proposed in this paper. Furthermore, RESOM is a real-time neural network that can adapt its learning parameters based on the scene behavior and it mimics perception abilities. A quantitative comparison with other segmentation methods reported in the literature using real video scenes showed that the proposed DR-SOM segmentation method automatically adjusts its parameters and outperforms the reported methods in condition of dynamic backgrounds, and gradual and sudden illumination changes.

The background modeling is the very first and essential part of every computer assisted surveillance system. Without it there would be no reliable way for fast and robust detection of moving objects in video sequences. In this paper we collect, describe and compare the main features of the most commonly used techniques of background modeling in video sequences and determine the most desirable way for the development of new algorithms in this field.

This paper presents a new statistical approach combining temporal and spatial information for robust background subtraction (BS) in videos. Temporal information couples finite mixtures of generalized Gaussians (MoGG) and temporal cooccurrence analysis of forground/background data. Spatial information combines multi-scale correlation analysis and histogram matching. Our approach fuses both information to perform efficient BS in the presence of shadows, illumination changes and various complex background dynamics. Comparison with recent state-of-the-art methods on standard datasets has demonstrated the performance of our method in terms of precision and computational efficiency.

<div><p><em>Tracking of moving objects in video sequences are essential for many computer vision applications & it is considered as a challenging research issue due to dynamic changes in objects, shape, complex background, illumination changes and occlusion. Many traditional tracking algorithms fails to track the moving objects in real-time, this paper proposes a robust method to overcome the issue, based on the combination of particle filter and Principal Component Analysis (PCA), which predicts the position of the object in the image sequences using stable wavelet features, which in turn are extracted from multi scale 2-D discrete wavelet transform. Later, PCA approach is used to construct the effective subspace. Similarity degree between the object model and the prediction obtained from particle filter is used to update the feature vector to handle occlusion and complex background in video frames. Experimental results obtained from the proposed method are encouraging.</em></p></div>

A noise robust method for 2D shape estimation of moving objects in video sequences considering a moving camera

A region-growing algorithm for detecting moving objects in video sequences based on optical flow is suggested; it calculates the optical flow only for moving objects and does not consider stationary image regions. The algorithm has linear complexity depending on the value of the optical flow window, uses SSE2/SSE3 instructions for significant acceleration of the calculations, and can be used in the real-time mode in problems of detection and tracking of moving objects in video sequences with fragmentary filling of frames with moving objects.

In the context of automotive driving assistance, this paper describes a generic (i.e. applicable to both vehicle interior and exterior scenes) vision based approach for scene content analysis. It makes use of temporal and spatial information from a stereoscopic sequence of images to localize objects and estimate their position and motion. The proposed method is divided into three steps. First, image features are selected, tracked and reconstructed in the 3D world space. Second, a clustering step is processed in the 5D space made of the positions and 2D motions parameters. The last step is devoted to clusters interpretation: it is out of the scope of the paper, however orientations are given to illustrate the capabilities of the proposed approach. The paper is organized as follows: first, the use of temporal and spatial information from a stereoscopic sequence is investigated. A state of the art of existing methods is presented. Then, a generic approach for object segmentation is proposed. Lastly, experimental results are presented.

The paper demonstrates a technique for analysing the following three problems. They are mainly associated with the automatic extraction of moving objects, suppression of the remaining errors, and tracking analysis under disturbed image situations from static camera. For this technique, a modified difference image-based approach for the segmentation of moving objects in video sequences is applied. The second part of the paper examines the problem of suppression of the remaining errors (holes, outliers or fusion of regions) by means of morphological and separation operators. The extracted image regions represent the object candidates for the following tracking. The efficiency of this suggested technology for moving objects segmentation will be demonstrated in this paper on the basis of the analysis of strongly disturbed image sequences.

In this paper we integrate spatial and temporal information, which are extracted separately from a video sequence, for indexing and retrieval purposes. We focus on two filter families that are suitable models of the human visual system for spatial and temporal information encoding. They are special cases of polynomial transforms that perform local decompositions of a signal. Spatial primitives are extracted using Hermite filters, which agree with the Gaussian derivative model of receptive field profiles. Temporal events are characterized by Laguerre filters, which preserve the causality constraint in the temporal domain. Integration of both models gives a spatio-temporal feature extractor based on early vision. They are efficiently implemented as two independent sets of discrete channels, Krawtchouk and Meixner, whose outputs are combined for indexing a video sequence. Results encourage our model for video indexing and retrieval.

This paper proposes a moving target detection scheme suitable for camera motion. Firstly, the background model is initialized by a Gaussian mixture model algorithm. Then Kanade-Lucas-Tomasi Feature Tracker (KLT) method is used to detect optical flow feature points of two adjacent frames, RANdom SAmple Consensus (RANSAC) algorithm is used to filter out the correct matching points and obtain a homography matrix, which can recover the background model matching the current frame, the new background model is used to detect moving target of the current frame. In the foreground detection stage, the current pixel is first compared with its own background model, and then compared with the background model of its 8 neighborhood pixels, the algorithm is speeded up without reducing the detection accuracy in this way; In the update stage of the background model, in order to adapt to the background changes caused by camera motion, an age value variable is set for each pixel. The experimental results show that the improved algorithm has a significant improvement in detection accuracy and running time compared to Gaussian mixture background modeling.

This paper presents a moving object detection method against sudden illumination change using improved background modeling. Initially, background model is created for every pixel from the first frame. The sample values for the model of a pixel are collected from the neighborhood of that pixel. Then the new pixel values from the new frames are compared to make background foreground decision. Conventional background modeling faces problem with change in the illumination of the scene. The proposed method frequently checks whether abrupt change of illumination take place or not and then initialize the background model from the frame that is detected with changed illumination. The illumination change is detected by obtaining the images of two frames that are taken at a suitable interval in HSV color space. Then the mean change value of each channel is calculated to make a decision. This enables the background model to start over with new sample values that are obtained in the current illumination condition and the background subtraction process can successfully detect moving object with greater accuracy even in changing illumination condition. Simulation results indicate that the proposed method gives excellent results in illumination changing condition to detect moving object whereas the conventional background modeling can not detect accurately. Comparison analysis shows that our proposed method outperforms recent methods in terms of detection accuracy.

There is context-related temporal and spatial information in video. How to make better use of these information is the key to improving the accuracy and speed of video object detection. The method of processing spatial and temporal information through 3D convolution is better than that of processing spatial information first and then processing time dimension information. There are two reasons for this assertion. One is that the 3D convolution can process spatial information and temporal information at the same time which will not divide them into two parts in feature extraction. The other is that the human visual derivation mechanism needs to memorize multi frames and process a video as a whole. After the training test on NFS dataset, it is verified that 3D convolution has stronger ability to extract temporal and spatial information. In addition, this paper also proposes a downward sampling method by connecting the low-resolution layer and the high-resolution layer, which can effectively improve the FPN model and enrich its high-level feature representation. After training and testing with different number of downward sample modules on MOT16 dataset, the improvement of model performance by down-sampling module is verified. Finally, this paper combines the downward sample module with the 3D network, and obtains the best test results on the NFS dataset.

We present a new statistical approach combining temporal and spatial information for robust online background subtraction (BS) in videos. Temporal information is modeled by coupling finite mixtures of generalized Gaussian distributions with foreground/background co-occurrence analysis. Spatial information is modeled by combining multiscale inter-frame correlation analysis and histogram matching. We propose an online algorithm that efficiently fuses both information to cope with several BS challenges, such as cast shadows, illumination changes, and various complex background dynamics. In addition, global video information is used through a displacement measuring technique to deal with pan-tilt-zoom camera effects. Experiments with comparison with recent state-of-the-art methods have been conducted on standard data sets. Obtained results have shown that our approach surpasses several state-of-the-art methods on the aforementioned challenges while maintaining comparable computational time.

This study presents a hierarchical background modelling and subtraction approach for real‐time detection of moving objects. At the first level, a novel pixel‐wise background modelling method is proposed for coarse detection. The method can dynamically assign the optimal number of components for each pixel with the borrow–lend strategy. And a flexible learning rate which is variable and different for each component is presented to adapt to scene changes. Additionally, a new mechanism using a framework of finite state machine is introduced to maintain and update the background models. At the second level, in order to deal with sudden illumination changes, a block‐wise foreground validation approach is adopted for refined detection. The authors compare the proposed approach with state‐of‐the‐art methods and experimental results under various scenes demonstrate the robustness and effectiveness of the proposed approach.

Moving object detection is an important component of a traffic surveillance system. Usual background subtraction approaches often poorly perform on a long outdoor traffic video due to vehicles waiting at an intersection and gradual changes of illumination and background shadow position. We present a fast and robust background subtraction algorithm based on unified spatio-temporal background and foreground modeling. The correlation between neighboring pixels provides high levels of detection accuracy in the dynamic background scene. Our Bayesian fusion method, which establishes the traffic scene model, combines both background and foreground models and considers prior probabilities to adapt changes of background in each frame. We explicitly model both temporal and spatial information based on the kernel density estimation (KDE) formulation for background modeling. Then, we use a Gaussian formulation to describe the spatial correlation of moving objects for foreground modeling. In the updating step, a fusion background frame is generated, and reasonable updating rates are also proposed for the traffic scene. The experimental results show that the proposed method outperforms the previous work with less computation and is better suited for the traffic scenes.