Illumination compensation based change detection using order consistency

Abstract

We present a change detection method resistant to global and local illumination variations for use in visual surveillance scenarios. Approaches designed thus far for robustness to illumination change are generally based either on color normalization, texture (e.g. edges, rank order statistics, etc.), or illumination compensation. Normalization based methods sacrifice discriminability while texture based methods cannot operate on texture-less regions. Both types of method can produce large missing regions in the distance image which in turn pose problems for higherlevel processing tasks that may be shape or region-based and require accurate foreground masks (e.g. person detection and tracking, crowd segmentation, etc.). Texture based methods have an additional problem in that they produce false alarms due to textures induced by local illumination effects (e.g. cast shadows). In this paper we propose a compensation based approach for change detection. Prior work on compensation has largely taken an empirical approach, and has not dealt with the important problem of rejecting outliers when they dominate the scene. In contrast, our generative approach and systematic handling of outliers enables us to achieve robustness to illumination change while eliminating the problems mentioned above. Furthermore, the computational complexity of our method is low enough for real-time performance. Results comparing images taken under strongly different illumination conditions, demonstrate the power and generality of the proposed method.

Linear colour correction for multiple illumination changes and non‐overlapping cameras

“…Nevertheless, this calculation is very sensitive to noise and introduces blurring artifacts in the image. Inspired by Grossberg and Nayar's work [12], Parameswaran [21] implemented an illumination compensation method based on the fact that the order of the pixel intensities is maintained when the illumination changes. When the illumination change is local, several functions must be estimated, which increases the computational cost.…”

An exposure guided background subtraction for smart camera

“…Another way to handle illumination change is background illumination compensation, which considers the illumination compensation as an inverse problem. They [6][7] first estimate the illumination change and then compensate the background illumination. However, these algorithms require high estimation accuracy and high computational cost to estimate the illumination change.…”

Change detection by local illumination compensation using local binary pattern

“…9 Because we assume the camera is usually static while conducting change detection, the geometrical distortion is ignored in this paper. [14][15][16][17] Therefore, we present a novel compositional method based on the combination of LBP and local illumination compensation (LIC). Illumination variation usually causes global and/or local illumination changes between images taken at different time instants.…”