WePBAS: A Weighted Pixel-Based Adaptive Segmenter for Change Detection

Abstract

The pixel-based adaptive segmenter (PBAS) is a classic background modeling algorithm for change detection. However, it is difficult for the PBAS method to detect foreground targets in dynamic background regions. To solve this problem, based on PBAS, a weighted pixel-based adaptive segmenter named WePBAS for change detection is proposed in this paper. WePBAS uses weighted background samples as a background model. In the PBAS method, the samples in the background model are not weighted. In the weighted background sample set, the low-weight background samples typically represent the wrong background pixels and need to be replaced. Conversely, high-weight background samples need to be preserved. According to this principle, a directional background model update mechanism is proposed to improve the segmentation performance of the foreground targets in the dynamic background regions. In addition, due to the “background diffusion” mechanism, the PBAS method often identifies small intermittent motion foreground targets as background. To solve this problem, an adaptive foreground counter was added to the WePBAS to limit the “background diffusion” mechanism. The adaptive foreground counter can automatically adjust its own parameters based on videos’ characteristics. The experiments showed that the proposed method is competitive with the state-of-the-art background modeling method for change detection.

Moving Vehicle Detection and Tracking Based on Video Sequences

“…To solve the above problems, a trust interval was set to judge if the current background model is suitable to update under the current traffic conditions. The interval was generated by the pixel-based adaptive segmentation algorithm [30], which automatically updates threshold R(x,y) and learning rate T(x,y) with the dynamic changes of the background. The background model B(x,y) can be defined as:…”

ABGS Segmenter: Pixel wise Adaptive Background Subtraction and Intensity Ratio based Shadow Removal Approach for Moving Object Detection in Video Surveillance Application

“…Comparison is done on the basis of two stages model. First stage is used to generate binary mask using background subtraction model and compare with ve state of the art methods (i) Temporal Average Filter [5], (ii) Running Average Gaussian (RAG) [22], (iii) Adaptive GMM [23], (iv) WePBAS [29], (v) ViBe [27]. These methods are chosen on the requirement of mapping threshold value for foreground segmentation.…”

“…Table 5 and Table 6 represent quantitative analysis of proposed method after post processing operations and observe that proposed method is able to generate better results than compared methods. There are few frames in Highway dataset which are associated with shadow due to which AGMM [23], WePBAS [29], and ViBe [27] methods generate more pixels misclassi cations (considering shadow pixels as foreground) and increases more false positive rate (FPR). It is observed that in Highway dataset, RAG [22] shows bad recall when more pixels are misclassi ed due to long shadow.…”

“…Another method called Pixel Based Adaptive Segmenter(PBAS) is developed which creates background model taking sample of observed pixels and foreground was detected using pixel based threshold value [28]. To extend this work, paper [29] proposes weighted directional background subtraction model to improve the segmentation task in static and dynamic scene.…”