Abstract:Background subtraction is one of the most fundamental and challenging tasks in computer vision. Many background subtraction algorithms work well under the assumption that the backgrounds are static over short time periods but degrade dramatically in dynamic scenes, such as swaying trees, rippling water, and waving curtains. In this paper, we propose an effective background subtraction method to address these difficulties by combining color features with texture features in the ViBe framework. Specifically, we present a novel local compact binary count (LCBC) feature that can capture local binary gray-scale difference information and totally discard the local binary structural information. The effective fusion of color and LCBC information significantly improves the performance of the ViBe model, making it very robust to background variations while still highlighting the moving objects. We further embed the total variation (TV) norm regularization technique into the proposed method, which can enhance the spatial smoothness of foreground objects, thereby further improving the accuracy of the method. We evaluate the proposed method against ten sequences containing dynamic backgrounds and show that our method outperforms many state-of-the-art methods in reducing the false positives without compromising the reasonable foreground definitions. The experimental results on challenging well-known data sets demonstrate that the proposed method works effectively on a wide range of dynamic background scenes.

Foreground Model for Background Subtraction with Blind Updating

New Method Of Background Update For Video-Based Vehicle Detection

Background Subtraction Based on Nonparametric Bayesian Estimation.

Background Subtraction with Dynamic Noise Sampling and Complementary Learning

Block Background Subtraction Method Based on ViBe

A nonlinear background updating scheme

A Novel Background Subtraction Method Based on ViBe.

Background subtraction via time continuity and texture consistency constraints.

Fast Background Subtraction Algorithm By Statistical Background Updating

Foreground Detection Based on Modified ViBe in Dynamic Background

An Improved Sample-based Model for Background Subtraction

Background Modeling Method for Moving Object Segmentation

Adaptive Difference Modelling for Background Subtraction.

Background Subtraction Using Incremental Subspace Learning

A New Background Update Algorithm for Airborne Camera in Dynamic Background

Complex Background Subtraction by Pursuing Dynamic Spatio-Temporal Models

Multi-Channel Adaptive Mixture Background Model for Real-time Tracking.

Background Subtraction Through Multiple Life Span Modeling

The Integration Adjacent Frame Difference of Improved ViBe for Foreground Object Detection

Local Compact Binary Count Based Nonparametric Background Modeling for Foreground Detection in Dynamic Scenes

New Background Subtraction Algorithm Based on Statistic