Automatic Detection and Tracking of Moving Objects in
Complex Environments for Video Surveillance Applications
ABSTRACT:
Autonomous video surveillance and monitoring has a rich history. A new method for
detecting multiple moving objects based on improved background subtraction model and for
tracking is based on feature based approach has proposed. Then identified moving objects are
also counted, by indexing individually. The proposed algorithm is automatic and efficient in
intelligent surveillance applications like vehicles monitoring, event recognition, and crime
prevention, etc. The proposed model has proved to be robust in various environments
(including indoor and outdoor scenes) and different types of background scenes. Experiments
on real scenes show that the algorithm is effective for object detection and tracking.
Key-Words: Object Detection, Multiple Tracking, Indexing, and Surveillance.
INTRODUCTION:
The capability of extracting moving objects from a video sequence is a fundamental and
crucial problem of many vision systems that include video surveillance, traffic monitoring,
human detection and tracking for video teleconferencing or human-machine interface, video
editing, and other applications. Typically, the usual approach for discriminating moving object
from the background scene is background subtraction. The idea of background subtraction is to
subtract the current image from a reference image, which separates the object of interest from
unrelated background, but contains scattering noise. The technique has been used for years in
many vision systems as a preprocessing step for object detection and tracking, . As per the
literature Survey, many of these algorithms are susceptible to both global and local illumination
changes such as shadows and highlights. This will cause the consequent processes of tracking,
recognizing, and also affects the accuracy and efficiency of the moving object. This problem is
the underlying motivation of proposed work. Tracking and indexing moving object is a method
to track down a single or multiple moving objects within a given environment. This is to isolate
the object within the image view without losing moving object. In the case of moving people, it
can differentiate between any kinds of actions within the camera range. All of the moving
object actions such as walking faster or slower, people carrying other objects or baby can be
recognized easily without error.
VEDLABS, #112, Oxford Towers, Old airport Road, Kodihalli, Bangalore-08
www.vedlabs.com , Email id: projects@vedlabs.com, Ph: 080-42040494.
Page 1
PROPOSED METHOD:
The concept of Background Subtraction and Feature Based approach is discussed.
A. Background Model:
To achieve moving object detection, there are many methods or models which are widely used
by researchers. Most widely used is background subtraction method. Background Subtraction
model works efficiently, since surveillance cameras are fixed and these are monitored towards
a fixed direction. The process works by taking the difference between the consecutive frames.
B. Preprocessing:
After the successful completion of background subtraction, little noise will be still encountered
which cannot be considered as moving object. To remove this kind of a noise, component
having box of height less than threshold.
C. Tracking and Indexing of Objects:
In Tracking and Indexing of moving objects, a feature based model will take any of the features
of the moving object image such as the coordination, size of the pixels, and distance from
current frame to the next frame or any other features which can be related to tracking down
the moving object. In order to achieve the best accuracy in tracking and recognizing each of the
moving objects the three features considered here are, Area, Centroids and Averages of RGB
pixels.
HARDWARE AND SOFTWARE REQUIREMENTS:
Software Requirement Specification:

Operating System: Windows XP with SP2

Tool: MATLAB R2010, Version: 7.10.0
Hardware Requirement specification:

Minimum Intel Pentium IV Processor

Primary memory: 2 GB RAM,
VEDLABS, #112, Oxford Towers, Old airport Road, Kodihalli, Bangalore-08
www.vedlabs.com , Email id: projects@vedlabs.com, Ph: 080-42040494.
Page 2
REFERENCES:
[1] Ismail Haritaoglu, David Harwood, and Larry S. Davis, W4: Who? When? Where? What? a
“Real-time System for Detecting and Tracking People," Proc. the third IEEE International
Conference on Automatic Face and Gesture Recognition (Nara, Japan), IEEE Computer Society
Press, Los Alamitos, Calif., pp.222-227,1998.
[2] .P.L.Rosin, “Thresholding for Change Detection," Proceedings of International Conference on
Computer Vision, 1998.
[3]. N. Friedman, and S. Russell, \Image Segmentation in Video Sequences: A Probabilistic
Approach", Proceedings of the 13th Conference on Uncertainty in Artificial intelligence, Morgan
Kaufmann, 1997.
[4]. C.R. Wren, A. Azarbayejani , T. Darrell, and A. Pentland, Pfinder: ‘’Real-time Tracking of the
Human Body," IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol.19, No.7, IEEE
Computer Society Press, Los Alamitos, Calif., July 1997, pp.780-785.
[5]. J. Ohya, et al., “Virtual Metamorphosis", IEEE Multimedia, April-June 1999.
VEDLABS, #112, Oxford Towers, Old airport Road, Kodihalli, Bangalore-08
www.vedlabs.com , Email id: projects@vedlabs.com, Ph: 080-42040494.
Page 3