Improved Gaussian Mixtures for Robust Object Detection by
Adaptive Multi-Background Generation
Mahfuzul Haque, Manzur Murshed, and Manoranjan Paul
Gippsland School of Information Technology, Monash University, Victoria 3842, Australia
Email: {Mahfuzul.Haque, Manzur.Murshed, Manoranjan.Paul}@infotech.monash.edu.au

1

Abstract
Adaptive Gaussian mixtures are widely used to model the dynamic background for real-time object detection. However,
object quality still remains unacceptable due to poor Gaussian mixture quality, susceptibility to background/foreground
data proportion, and instability with varying operating environments. This paper presents an effective technique to
eliminate these drawbacks by modifying the new model induction logic and using intensity difference thresholding to
detect objects from one or more believe-to-be backgrounds.

2 Background Modelling

6 Quantitative Evaluation

Frame 1
Frame 2
..
Frame t

Input scenes
Experimental results on 14 test
sequences including PETS and
Wallflower datasets.

Gaussian Mixture Model (GMM)
for each pixel

Error rates at medium learning
rate (α = 0.01) and the standard
deviation of the error rates over
three learning rates α = 0.1, α =
0.01, and α = 0.001.

P( X t )  iK 1 wi,t  ( X t ,  i,t ,  i,t )

 ( X t , t ,  t ) 

1
(2 )

n/2

||

1/ 2

e

1
 ( X t   t )T
2

 1 ( X t  t )

A pixel model is constructed and updated for each pixel which maintains a
mixture of Gaussian distributions for modelling multi-modal distribution
caused by moving foregrounds and repetitive background motions [1-3].

7 Qualitative Evaluation
First
Frame

3 New Model Induction Scheme
New Model

Test
Frame

Ideal
Result

Lee’s
Tech.

Proposed
Tech.

Existing Models

P(x)

Intensity

4 Proposed Detection Scheme
Model matching:

Visual comparison results at medium learning rate, α = 0.01.

B/G Model Selection:

;

8

Implemented System

F/G Detection:

5 Model Quality Visualisation
Model distance

0

127

255

One model
Two models
More than two models

Input Frames

Visualisation

The images shown in the header has been taken
from http://www.informationliberation.com

[1] Mahfuzul Haque, Manzur Murshed, and Manoranjan Paul, On Stable Dynamic Background Generation Technique using
Gaussian Mixture Models for Robust Object Detection, IEEE International Conference On Advanced Video and Signal Based
Surveillance (AVSS), New Mexico, USA, 2008.
[2] Mahfuzul Haque, Manzur Murshed, and Manoranjan Paul, A Hybrid Object Detection Technique from Dynamic
Background Using Gaussian Mixture Models, IEEE International Workshop on Multimedia Signal Processing (MMSP), Cairns,
Australia, 2008.
[3] Mahfuzul Haque, Manzur Murshed, and Manoranjan Paul, Improved Gaussian Mixtures for Robust Object Detection by
Adaptive Multi-Background Generation, International Conference on Pattern Recognition (ICPR), Tampa, Florida, USA, 2008.

Poster: ICPR 2008

  • 1.
    Improved Gaussian Mixturesfor Robust Object Detection by Adaptive Multi-Background Generation Mahfuzul Haque, Manzur Murshed, and Manoranjan Paul Gippsland School of Information Technology, Monash University, Victoria 3842, Australia Email: {Mahfuzul.Haque, Manzur.Murshed, Manoranjan.Paul}@infotech.monash.edu.au 1 Abstract Adaptive Gaussian mixtures are widely used to model the dynamic background for real-time object detection. However, object quality still remains unacceptable due to poor Gaussian mixture quality, susceptibility to background/foreground data proportion, and instability with varying operating environments. This paper presents an effective technique to eliminate these drawbacks by modifying the new model induction logic and using intensity difference thresholding to detect objects from one or more believe-to-be backgrounds. 2 Background Modelling 6 Quantitative Evaluation Frame 1 Frame 2 .. Frame t Input scenes Experimental results on 14 test sequences including PETS and Wallflower datasets. Gaussian Mixture Model (GMM) for each pixel Error rates at medium learning rate (α = 0.01) and the standard deviation of the error rates over three learning rates α = 0.1, α = 0.01, and α = 0.001. P( X t )  iK 1 wi,t  ( X t ,  i,t ,  i,t )   ( X t , t ,  t )  1 (2 ) n/2 || 1/ 2 e 1  ( X t   t )T 2  1 ( X t  t ) A pixel model is constructed and updated for each pixel which maintains a mixture of Gaussian distributions for modelling multi-modal distribution caused by moving foregrounds and repetitive background motions [1-3]. 7 Qualitative Evaluation First Frame 3 New Model Induction Scheme New Model Test Frame Ideal Result Lee’s Tech. Proposed Tech. Existing Models P(x) Intensity 4 Proposed Detection Scheme Model matching: Visual comparison results at medium learning rate, α = 0.01. B/G Model Selection: ; 8 Implemented System F/G Detection: 5 Model Quality Visualisation Model distance 0 127 255 One model Two models More than two models Input Frames Visualisation The images shown in the header has been taken from http://www.informationliberation.com [1] Mahfuzul Haque, Manzur Murshed, and Manoranjan Paul, On Stable Dynamic Background Generation Technique using Gaussian Mixture Models for Robust Object Detection, IEEE International Conference On Advanced Video and Signal Based Surveillance (AVSS), New Mexico, USA, 2008. [2] Mahfuzul Haque, Manzur Murshed, and Manoranjan Paul, A Hybrid Object Detection Technique from Dynamic Background Using Gaussian Mixture Models, IEEE International Workshop on Multimedia Signal Processing (MMSP), Cairns, Australia, 2008. [3] Mahfuzul Haque, Manzur Murshed, and Manoranjan Paul, Improved Gaussian Mixtures for Robust Object Detection by Adaptive Multi-Background Generation, International Conference on Pattern Recognition (ICPR), Tampa, Florida, USA, 2008.