Motion Saliency Detection for Surveillance Systems Using Streaming Dynamic Mode Decomposition
Open Access
Abstract
Intelligent surveillance systems enable secured visibility features in the smart city era. One of the major models for pre-processing in intelligent surveillance systems is known as saliency detection, which provides facilities for multiple tasks such as object detection, object segmentation, video coding, image re-targeting, image-quality assessment, and image compression. Traditional models focus on improving detection accuracy at the cost of high complexity. However, these models are computationally expensive for real-world systems. To cope with this issue, we propose a fast-motion saliency method for surveillance systems under various background conditions. Our method is derived from streaming dynamic mode decomposition (s-DMD), which is a powerful tool in data science. First, DMD computes a set of modes in a streaming manner to derive spatial–temporal features, and a raw saliency map is generated from the sparse reconstruction process. Second, the final saliency map is refined using a difference-of-Gaussians filter in the frequency domain. The effectiveness of the proposed method is validated on a standard benchmark dataset. The experimental results show that the proposed method achieves competitive accuracy with lower complexity than state-of-the-art methods, which satisfies requirements in real-time applications.Keywords
This publication has 45 references indexed in Scilit:
- VSI: A Visual Saliency-Induced Index for Perceptual Image Quality AssessmentIEEE Transactions on Image Processing, 2014
- Salient object detection via local saliency estimation and global homogeneity refinementPattern Recognition, 2014
- Saliency-Aware Video CompressionIEEE Transactions on Image Processing, 2013
- Temporal Spectral Residual for fast salient motion detectionNeurocomputing, 2012
- A Spatiotemporal Saliency Model for Video SurveillanceCognitive Computation, 2011
- Dynamic mode decomposition of numerical and experimental dataJournal of Fluid Mechanics, 2010
- SUN: A Bayesian framework for saliency using natural statisticsJournal of Vision, 2008
- Predicting visual fixations on video based on low-level visual featuresVision Research, 2007
- Components of bottom-up gaze allocation in natural imagesVision Research, 2005
- A model of saliency-based visual attention for rapid scene analysisIEEE Transactions on Pattern Analysis and Machine Intelligence, 1998