Moving Object Detection: A New Approach
01 Jan 2014-pp 25-48
TL;DR: This chapter presents a novel background subtraction technique for detecting moving objects from video under dynamic background conditions that would be useful particularly for applications in resource constrained environments by virtue of its low computation time and storage requirements.
Abstract: This chapter presents a novel background subtraction technique for detecting moving objects from video under dynamic background conditions. The presented methodology is a simple and low-cost solution for modeling and updating background during a background subtraction process. Comparative performance analysis with other state-of-the-art methods on benchmark data-set shows the effectiveness of the proposed method. The objective of the research is not to claim that the proposed method yields the best result in terms of accuracy; rather the main novelty is in its low computational cost. However, in spite of being less computation-intensive, comparative analysis reveals that the new method produces quite competitive results as compared to other methods. The proposed method would be useful particularly for applications in resource constrained environments by virtue of its low computation time and storage requirements.
Citations
More filters
TL;DR: This paper provides a review of the human motion detection methods focusing on background subtraction technique and concludes that current methods for detecting objects in motion within videos from static cameras are inadequate.
Abstract: For the majority of computer vision applications, the ability to identify and detect objects in motion has become a crucial necessity. Background subtraction, also referred to as foreground detection is an innovation used with image processing and computer vision fields when trying to detect an object in motion within videos from static cameras. This is done by deducting the present image from the image in the background or background module. There has been comprehensive research done in this field as an effort to precisely obtain the region for the use of further processing (e.g. object recognition). This paper provides a review of the human motion detection methods focusing on background subtraction technique.
20 citations
Additional excerpts
...The Performance Evaluation of Tracking and Surveillance (PETS2006) is in charge of providing the benchmark surveillance dataset that was used for examining the execution of the suggested method [1,11] ....
[...]
TL;DR: The proposed method uses a combination of histogram specification and iterative histogram partitioning to progressively adjust the dynamic range and efficiently suppress the background of each video frame to ensure that pedestrians are present in the image at the convergence point of the algorithm.
Abstract: This paper presents a novel candidate generation algorithm for pedestrian detection in infrared surveillance videos. The proposed method uses a combination of histogram specification and iterative histogram partitioning to progressively adjust the dynamic range and efficiently suppress the background of each video frame. This pairing eliminates the general-purpose nature associated with histogram partitioning where chosen thresholds, although reasonable, are usually not suitable for specific purposes. Moreover, as the initial threshold value chosen by histogram partitioning is sensitive to the shape of the histogram, specifying a uniformly distributed histogram before initial partitioning provides a stable histogram shape. This ensures that pedestrians are present in the image at the convergence point of the algorithm. The performance of the method is tested using four publicly available thermal datasets. Experiments were performed with images from four publicly available databases. The results show the improvement of the proposed method over thresholding with minimum-cross entropy, the robustness across images acquired under different conditions, and the comparable results with other methods in the literature.
3 citations
TL;DR: A spatio-temporal processing scheme to improve automatic vehicle detection performance by replacing the thresholding step of existing detection algorithms with multi-neighborhood hysteresis thresholding for foreground pixel classification is presented.
Abstract: Vehicle detection in aerial videos often requires post-processing to eliminate false detections. This paper presents a spatio-temporal processing scheme to improve automatic vehicle detection performance by replacing the thresholding step of existing detection algorithms with multi-neighborhood hysteresis thresholding for foreground pixel classification. The proposed scheme also performs spatial post-processing, which includes morphological opening and closing to shape and prune the detected objects, and temporal post-processing to further reduce false detections. We evaluate the performance of the proposed spatial processing on two local aerial video datasets and one parking vehicle dataset, and the performance of the proposed spatio-temporal processing scheme on five local aerial video datasets and one public dataset. Experimental evaluation shows that the proposed schemes improve vehicle detection performance for each of the nine algorithms when evaluated on seven datasets. Overall, the use of the proposed spatio-temporal processing scheme improves average F-score to above 0.8 and achieves an average reduction of 83.8% in false positives.
2 citations
Cites background from "Moving Object Detection: A New Appr..."
...PWC [6], [14], [27]–[30] is defined as the ratio of false detections andmissed objects to the sum of detections and missed objects, which is formulated as [6], [30]...
[...]
...detection and segmentation in aerial surveillance are mainly categorized as frame differencing, background subtraction, and optical flow approaches [14], [19], [20], each having advantages and disadvantages [20]....
[...]
TL;DR: In this paper, the authors proposed a method for efficient extraction of human silhouette from video sequences using background elimination, edge detection, region filling and noise removal using morphological operations to estimate the silhouette of an image.
Abstract: In this paper we propose a method for efficient extraction of human silhouette from video sequences. The proposed approach includes background elimination, edge detection, region filling and noise removal using morphological operations to estimate the silhouette of an image. To the best of our knowledge our proposed approach for silhouette extraction involving background elimination and edge detection is first of its kind. We have applied our proposed technique on Weizmann (standard) dataset and compared the results with the most recent related research work. The comparison results in terms of statistical measures like precision, recall and F-measure clearly show the supremacy of our method and thus justify its novelty.
References
More filters
16 Jun 2012
TL;DR: In this paper, a novel method for foreground segmentation is presented that follows a non-parametric background modeling paradigm, thus the background is modeled by a history of recently observed pixel values and the background update is based on a learning parameter.
Abstract: In this paper we present a novel method for foreground segmentation. Our proposed approach follows a non-parametric background modeling paradigm, thus the background is modeled by a history of recently observed pixel values. The foreground decision depends on a decision threshold. The background update is based on a learning parameter. We extend both of these parameters to dynamic per-pixel state variables and introduce dynamic controllers for each of them. Furthermore, both controllers are steered by an estimate of the background dynamics. In our experiments, the proposed Pixel-Based Adaptive Segmenter (PBAS) outperforms most state-of-the-art methods.
583 citations
28 Mar 2013
TL;DR: A new background model considering the similarity in the intensity changes among pixels is proposed, which classifies all the pixels into several clusters based on the similarity of their intensity changes.
Abstract: Object detection is an important task for computer vision applications. Many researchers have proposed a lot of methods to detect the objects through the background modeling. Most of previous approaches model the background independently for each pixel and detect foreground objects based on it. Then, it is difficult for the background model to deal with illumination changes, which cause significant intensity changes as in the case that a foreground object appears. To solve this problem, in this paper, we propose a new background model considering the similarity in the intensity changes among pixels. In particular, we classify all the pixels into several clusters based on the similarity of their intensity changes. Then, focusing on each cluster, we can easily identify whether the significant intensity changes are caused by foreground objects or illumination changes. This is because, if the illumination changes, most of the pixels belonging to the same cluster exhibit the similar intensity changes.
22 citations