International Journal of Electrical and Computer Engineering (IJECE) Vol. 8, No. 5, October 2018, pp. 2847~2856 ISSN: 2088-8708, DOI: 10.11591/ijece.v8i5.pp2847-2856  2847 Journal homepage: http://iaescore.com/journals/index.php/IJECE Background Estimation Using Principal Component Analysis Based on Limited Memory Block Krylov Subspace Optimization Ilmiyati Sari 1 , Asep Juarna 2 , Suryadi Harmanto 3 , Djati Kerami 4 1,2,3 Department of Computer Science and Information Technology, Gunadarma University, Indonesia 4 Department of Mathematics, University of Indonesia, Indonesia Article Info ABSTRACT Article history: Received Nov 16, 2017 Revised Jan 12, 2018 Accepted Apr 1, 2018 Given a video of  frames of size ℎ × . Background components of a video are the elements matrix which relative constant over  frames. In PCA (principal component analysis) method these elements are referred as “principal components”. In video processing, background subtraction means excision of background component from the video. PCA method is used to get the background component. This method transforms 3 dimensions video (ℎ ×  × ) into 2 dimensions one ( × ), where  is a linear array of size ℎ × . The principal components are the dominant eigenvectors which are the basis of an eigenspace. The limited memory block Krylov subspace optimization then is proposed to improve performance the computation. Background estimation is obtained as the projection each input image (the first frame at each sequence image) onto space expanded principal component. The procedure was run for the standard dataset namely SBI (Scene Background Initialization) dataset consisting of 8 videos with interval resolution [146 150, 352 240], total frame [258,500]. The performances are shown with 8 metrics, especially (in average for 8 videos) percentage of error pixels (0.24%), the percentage of clustered error pixels (0.21%), multiscale structural similarity index (0.88 form maximum 1), and running time (61.68 seconds). Keyword: Background estimation Eigen vector Krylov PCA Subspace optimization Copyright © 2018 Institute of Advanced Engineering and Science. All rights reserved. Corresponding Author: Ilmiyati Sari, Department of Computer Science and Information Technology, Gunadarma University, Margonda Raya Road, Pondok Cina, Depok, West Java 16424, Indonesia. Email: ilmiyati.sari25@gmail.com 1. INTRODUCTION Background subtraction is an important step in many computer vision systems to detect moving objects [1]. It is commonly used in video surveillance applications to detect persons, vehicles, animals, etc., before operating more complex processes for intrusion detection, tracking, people counting, etc. The basic operation consists of separating the moving objects called “foreground” from the static information called “background”[2]. It consists in using a model of the scene background in order to detect foreground objects by differencing incoming frames with the model. Indeed, the first step in background subtraction is background estimation. We state the general problem of background estimation, also known as background initialization, bootstrapping, background reconstruction, initial background extraction, or background generation, as follows: Given a set of images of a scene taken at different times, in which the background is occluded by any number of foreground objects, the aim is to determine a model describing the scene background with no foreground objects [3].