Abstract:Background subtraction is crucial for detecting moving objects in videos. However, most background subtraction methods usually fail to deal well with illumination changes, static foreground, ghosts, and dynamic background. In this paper, we propose an efficient coarse-to-fine sample-based algorithm to address these issues simultaneously. Specifically, a rough motion region estimation method, is first proposed by combining an improved frame difference method, the block-based image partition algorithm and multi-scale region-based technique to locate the possible area of moving objects. It can compensate the global illumination change and suppress dynamic background. Then, an improved Vibe algorithm with adaptive distance threshold and time sub-sampling factor parameters for each pixel is proposed to detect precise moving objects. It can not only retain static foreground for a long time while eliminating ghosts rapidly, but also detect camouflaged foreground efficiently. Experimental results demonstrate that our proposed algorithm can extract moving objects more accurately with less time cost than several state-of-the-art background subtraction approaches.

A new background subtraction method using texture and color information

Online background subtraction based on sufficient statistics

A Moving Objects Tracking Method Based on a Combination of Local Binary Pattern Texture and Hue

Dynamic Background Subtraction Using Spatial-Color Binary Patterns

Background subtraction via time continuity and texture consistency constraints.

Detection Algorithm of Moving Objects Based on Background Subtraction Method

New Background Subtraction Algorithm Based on Statistic

A Moving Objects Detection Method Based on a Combination of Improved Local Binary Pattern Texture and Hue

Moving Objects Detection Based on RGB Color Space and Background Subtraction

Texture And Motion Pattern Fusion For Background Subtraction

Coarse-to-fine Sample-Based Background Subtraction for Moving Object Detection

Moving Objects Detection Method Based on Brightness Distortion and Chromaticity Distortion

Motion-Based Background Subtraction

Background subtraction using shape and colour information

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

HLBP model method with color and location information about moving objects detection

Dynamic Background Subtraction Based on Spatial Extended Center-Symmetric Local Binary Pattern

Foreground Detection Based on Color and Texture Features

Detection of Moving Objects with Fuzzy Color Coherence Vector

A Fuzzy Background Model For Moving Object Detection

Dynamic background subtraction based on appearance and motion pattern