Object Classification and Tracking in Real Time: An Overview

Object Classification and Tracking in Real Time: An Overview

Amlan Jyoti Das (Gauhati University, India), Navajit Saikia (Assam Engineering College, India) and Kandarpa Kumar Sarma (Gauhati University, India)
DOI: 10.4018/978-1-4666-9685-3.ch011
OnDemand PDF Download:
$37.50

Abstract

Algorithms for automatic processing of visual data have been a topic of interest since the last few decades. Object tracking and classification methods are highly demanding in vehicle traffic control systems, surveillance systems for detecting unauthorized movement of vehicle and human, mobile robot applications, animal tracking, etc. There are still many challenging issues while dealing with dynamic background, occlusion, etc. in real time. This chapter presents an overview of various existing techniques for object detection, classification and tracking. As the most important requirements of tracking and classification algorithms are feature extraction and selection, different feature types are also included.
Chapter Preview
Top

1. Object Detection

Detection of object of interest in video is the most important step in a system for tracking and classifying object. The performance of the complete system depends on the accuracy and rate of detection of the region of interest. An object can be detected either when it first appears in the video or in every frame depending on requirement. Since it is more meaningful to detect the moving objects than to detect all the static objects in a video sequence, most methods focus on detecting such objects. The common approach is to use the temporal information that highlights the difference between the consecutive frames in a video. In the following, some popular object detection techniques are discussed in the context of object classification and tracking. The different types of object detection methods commonly used may be grouped as in Figure 1.

Figure 1.

Classification of Object detection techniques

Complete Chapter List

Search this Book:
Reset