A Real-time Object Detection System Using Selected Principal Components

The detection of moving objects is a basic and necessary preprocessing step in many applications such as object recognition, context awareness, and intelligent visual surveillance. Among these applications, object detection for context awareness impacts t

  • PDF / 1,835,589 Bytes
  • 10 Pages / 439.37 x 666.142 pts Page_size
  • 43 Downloads / 249 Views

DOWNLOAD

REPORT


Abstract The detection of moving objects is a basic and necessary preprocessing step in many applications such as object recognition, context awareness, and intelligent visual surveillance. Among these applications, object detection for context awareness impacts the efficiency of the entire system and it requires rapid detection of accurate shape information, a challenge specially when a complicated background or a background change occurs. In this paper, we propose a method for detecting a moving object rapidly and accurately in real time when changes in the background and lighting occur. First, training data collected from a background image are linearly transformed using principal component analysis (PCA). Second, an eigen-background is organized from selected principal components with excellent ability to discriminate between object and background. Finally, an object is detected by convoluting the eigenvector organized in the previous step with an input image, the result of which is the input value used on an EM algorithm. An image sequence that includes various moving objects at the same time is organized and used as training data to realize a system that can adapt to changes in J.-H. Kim  H.-S. Kim  S.-K. Kim (&) Department of Computer Engineering, Inje University, Gimhae, Gyeongsangnam-do 621-749, Republic of Korea e-mail: [email protected] J.-H. Kim e-mail: [email protected] H.-S. Kim e-mail: [email protected] B.-D. Kang Researcher, STAR Team, Korea Electronics Technology Institute, Bucheon-si, Gyeonggi-do 420-734, Republic of Korea e-mail: [email protected] S.-H. Ahn Department of Electronic Engineering, Inje University, Gimhae, Gyeongsangnam-do 621-749, Republic of Korea e-mail: [email protected]

J. J. (Jong Hyuk) Park et al. (eds.), Multimedia and Ubiquitous Engineering, Lecture Notes in Electrical Engineering 240, DOI: 10.1007/978-94-007-6738-6_46, Ó Springer Science+Business Media Dordrecht(Outside the USA) 2013

367

368

J.-H. Kim et al.

lighting and background. Test results show that the proposed method is robust to these changes, as well as to the partial movement of objects.



Keywords Object detection Principal Components Analysis (PCA) background Mixture of Gaussian



 Eigen-

1 Introduction Computer vision technology, which was originally developed for human computer interaction, has been applied to a variety of fields such as user interface designs, robot learning, and intelligent surveillance systems. Object detection, which accurately and effectively separates an object from its background, is an essential technology. Without proper foreground/background separation, it would be difficult to detect objects or analyze gestures in the next step of the system such as that required in vision-based robotic manipulation, augmented reality, and gesture recognition. Therefore, a number of researchers have been studying how to separate an object from its background. Representative methods utilize the difference between a previous frame or a previously saved background and the current frame [