Title :
Online background learning for illumination-robust foreground detection
Author :
Li, Dawei ; Xu, Lihong ; Goodman, Erik
Author_Institution :
Dept. of Control Sci. & Eng., Tongji Univ., Shanghai, China
Abstract :
This paper presents a background modeling algorithm and a foreground detecting method which is robust against illumination change, providing a novel and practical choice for intelligent video surveillance systems using static cameras. This paper first introduces an online Expectation Maximization algorithm which is developed from the basic batch edition to update the mixture models in real time. Then a spherical K-means clustering method is used to provide more accurate direction for the update of Gaussian Mixture Models after a deep study of RGB space features under illumination changes. Foreground detection is carried out using a statistical framework and RGB pixel intensity judgments. The results show the proposed algorithm outcompete several classic methods in efficiency, accuracy, and robustness to perturbations from illumination changes, on a sampling of problems.
Keywords :
Gaussian processes; Internet; computer aided instruction; pattern clustering; video surveillance; Gaussian mixture models; RGB pixel intensity; foreground detecting method; illumination robust foreground detection; intelligent video surveillance systems; k-means clustering method; online background learning; online expectation maximization algorithm; static cameras; statistical framework; Adaptation model; Clustering algorithms; Computational modeling; Hidden Markov models; Image color analysis; Lighting; Pixel; Gaussian mixture model (GMM); background modeling; expectation maximization; foreground detection; maximum likelihood estimation (MLE); online spherical k-means clustering (OSKMC);
Conference_Titel :
Control Automation Robotics & Vision (ICARCV), 2010 11th International Conference on
Conference_Location :
Singapore
Print_ISBN :
978-1-4244-7814-9
DOI :
10.1109/ICARCV.2010.5707245