Abstract
Modern image processing applications not only induce huge computational load but also are characterized by increasing complexity. As exemplarily shown in Section 2.2, they typically consist of a mixture of static and data-dependent algorithms and operate on both one-dimensional and multidimensional streams of data. Efficient implementation is only possible by exploiting different kinds of parallelism, namely task, data, and operation-level parallelism (see Section 2.5). Out-of-order communication and sliding windows with parallel data access require complex communication synthesis.
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
Notes
- 1.
- 2.
In accordance with [165, 166, 215] proposed to use a diagonal sampling matrix in order to describe the window movement, because this permits to represent non-rectangular sampling patterns. However, since this will not be used in this book , the matrix is replaced by a vector in order to ease notation.
- 3.
Assuming edge buffers of infinite size.
- 4.
This block size has only be chosen for illustration purposes. In reality, blocks are much larger and their size has to be a power of 2.
- 5.
It could be imagined to relax this condition in order to support also incomplete code-blocks, which can occur in JPEG2000 compression for certain relations between code-block and image size. As this, however, complicates both analysis and synthesis, it is not further detailed.
- 6.
More precisely spoken, \(pos\left(\right)\) returns the value of the hierarchical iteration vector defined later on in Section 6.2. It describes the current window or token position by taking the communication order into account.
- 7.
If such a schedule sequence cannot be found, the number of data elements stored in the edge buffers has to change permanently. As they are, however, positive, the only possibility is a–-possibly slow–-convergence toward infinity.
- 8.
WSDF currently requires that virtual tokens are produced and consumed completely. In other words it is not allowed to process a virtual token only half.
- 9.
Due to implementation details, the current read or write position within an action is returned via the iteration method. On the other hand, in the definition of the communication state machine, the function getIteration has to be used.
References
Buck, J.T.: Scheduling dynamic dataflow graphs with bounded memory using the token flow model. Ph.D. thesis, University of California at Berkeley (1993)
ISO/IEC JTC1/SC29/WG1: JPEG2000 Part I Final Committee Draft Version 1.0 (2002). N1646R
Karp, R.M., Miller, R.E.: Properties of a model for parallel computations: Determinacy, termination and queuing. SIAM J. Appl. Math. 14(6), 1390–1411 (1966)
Keinert, J., Haubelt, C., Teich, J.: Windowed Synchronous Data Flow (WSDF). Tech. Rep. 02-2005, University of Erlangen-Nuremberg, Institut for Hardware-Software-Co-Design (2005)
Keinert, J., Haubelt, C., Teich, J.: Modeling and analysis of windowed synchronous algorithms. ICASSP2006 III, 892–895 (2006)
Lee, E.A., Messerschmitt, D.G.: Static scheduling of synchronous data flow programs for digital signal processing. IEEE Trans. Comput. C-36(1), 24–35 (1987)
Murthy, P.K., Lee, E.A.: Multidimensional synchronous dataflow. IEEE Trans. Signal Process. 50(7), 2064–2079 (2002)
Reiter, R.: Scheduling parallel computations. J. ACM 15(4), 590–599 (1968)
Richardson, I.E.G.: H.264 and MPEG-4 Video Compression – Video Coding for Next-generation Multimedia. Wiley, West Sussex, England (2003)
Vincent, L.: Morphological grayscale reconstruction in image analysis: applications and efficient algorithms. IEEE Trans. Image Process. 2(2), 176–201 (1993)
Zebelein, C., Falk, J., Haubelt, C., Teich, J.: Classification of general data flow actors into known models of computation. In: Proc. 6th ACM/IEEE International Conference on Formal Methods and Models for Codesign (MEMOCODE 2008), pp. 119–128. Anaheim, CA (2008)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2011 Springer Science+Business Media, LLC
About this chapter
Cite this chapter
Keinert, J., Teich, J. (2011). Windowed Data Flow (WDF). In: Design of Image Processing Embedded Systems Using Multidimensional Data Flow. Embedded Systems. Springer, New York, NY. https://doi.org/10.1007/978-1-4419-7182-1_5
Download citation
DOI: https://doi.org/10.1007/978-1-4419-7182-1_5
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4419-7181-4
Online ISBN: 978-1-4419-7182-1
eBook Packages: EngineeringEngineering (R0)