Abstract
Previous research has shown that Fast-Fourier-Transform based method was an effective approach for studying computational attention model. In this paper, a quantitative analysis was carried out to explore the intrinsic mechanism of FFT-based approach. Based on it, a unified framework was proposed to summarize all existing FFT-based attention models. A new saliency detective model called Frequency Spectrum Modification (FSM) was also derived from this framework. Multiple feature channels and lateral competition were applied in this model for simulating human visual system. The comparison between FSM and other FFT-based models was implemented by comparing their responses with the real human eye’s fixation traces. The result leads to the conclusion that FSM is more effective in saliency detection.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Koch, C., Ullman, S.: Shifts in selective visual-attention – towards the underlying neural circuitry. Human Neurobiology 4, 219–227 (1985)
Itti, L., Koch, C., Niebur, E.: A model of saliency-based visual attention for rapid scene analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence 20(11), 1254–1259 (1998)
Walther, D., Koch, C.: Modeling attention to salient proto-objects. Neural Networks 19, 1395–1407 (2006)
Chen, D.Y., Zhang, L.M., Weng, J.: Spatiotemporal Adaptation in the Unsupervised Development of Networked Visual Neurons. IEEE Trans. on Neural Networks 20(6), 992–1008 (2009)
Hou, X., Zhang, L.: Saliency Detection: A Spectral Residual Approach. In: Proc. CVPR (2007)
Guo, C.L., Ma, Q., Zhang, L.M.: Spatio-temporal Saliency Detection Using Phase Spectrum of Quaternion Fourier Transform. In: Proc. CVPR (2008)
Ruderman, D.: The Statistics of Natural Images. Network: Computation in Neural Systems 5(4), 517–548 (1994)
Bruce, N.D., Tsotsos, J.K.: Saliency based on Information Maximization. In: Proc. NIPS (2005)
Chen, D.Y., Wu, C.D.: A New Model of Visual Attention Selection Based on Amplitude Modulation Fourier Transform. In: CCDC 2009 (2009)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Chen, D., Han, P., Wu, C. (2010). Frequency Spectrum Modification: A New Model for Visual Saliency Detection. In: Zhang, L., Lu, BL., Kwok, J. (eds) Advances in Neural Networks - ISNN 2010. ISNN 2010. Lecture Notes in Computer Science, vol 6064. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-13318-3_12
Download citation
DOI: https://doi.org/10.1007/978-3-642-13318-3_12
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-13317-6
Online ISBN: 978-3-642-13318-3
eBook Packages: Computer ScienceComputer Science (R0)