Multi images retrieved for automatic intruder detection using kernel-based learning algorithm

Abstract

Over the course of the last few decades, security systems have undergone radical overhauls and have transitioned into such sophisticated devices that some even completely overcome any form of human interventions. In this paper we apply a recursive Kernel-based Online Anomaly Detection algorithm to propose an automated, real-time intruder detection mechanism for surveillance networks. Our proposed method is portable and adaptive, and does not require any expensive or advanced components. Real images are collected using a rotating camera over a span of space and time, along with the comparison using common methods based on Principle Component Analysis (PCA), can show that it is possible to obtain high detection accuracy with low complexity. Thus, it could also exhibit sensitivity to threshold choices, and have no natural distributed form.