Incremental Eigenspace Model Applied to Monitoring System

This paper describes a real time feature extraction for real-time surveillance system. We use incremental KPCA method in order to represent images in a low-dimensional subspace for real-time surveillance in the traditional approach to calculate these eigen space models, known as batch PCA method, model must capture all the images needed to build the internal representation. Updating of the existing eigen space is only possible when all the images must be kept in order to update the eigen space, requiring a lot of storage capability. Proposed method allows discarding the acquired images immediately after the update. By experimental results we can show that incremental KPCA has similar accuracy compare to KPCA and more efficient in memory requirement than KPCA. This makes pro-posed model is suitable for real time surveillance system. We will extend our research to real time face recognition based on this research.