A maximum entropy Kalman filter for image compression

TitleA maximum entropy Kalman filter for image compression
Publication TypeConference Paper
Year of Publication2000
AuthorsDavid, A., and T. Aboulnasr
Conference NameCircuits and Systems, 2000. Proceedings of the 43rd IEEE Midwest Symposium on
Pagination884 -887 vol.2
Keywordsdata compression, digital image compression, edge detection, image coding, image texture, Kalman filters, maximum entropy Kalman filter, maximum entropy methods, optimisation, optimization

In this paper, we propose a novel compression method applicable to digital images. We employ Maximum Entropy (ME) as the optimization criterion and Kalman Filter (KF) as means of implementing the compressor. We will show for compression ratios comparable to those of traditional methods, such as JPEG, the high frequency components of the signal, i.e. texture and edges, are preserved. The motivation for using ME as the optimization criterion is to avoid over-smoothing of the signal associated with traditional methods based on Mean Square Error (MSE). The ME criterion is motivated by the fact that it does not make any assumptions, regarding the unobserved data


a place of mind, The University of British Columbia

Electrical and Computer Engineering
2332 Main Mall
Vancouver, BC Canada V6T 1Z4
Tel +1.604.822.2872
Fax +1.604.822.5949

Emergency Procedures | Accessibility | Contact UBC | © Copyright 2021 The University of British Columbia