Abstract
Symbolic grouping problems are key intermediate-level vision problems. These problems are characterized by uneven distribution of symbolic features among the processors, unbalanced workload, and irregular inter-processor data dependencies caused by the input image. In this paper, we propose an asynchronous algorithm for solving symbolic grouping problems on distributed memory machines. The algorithm balances the workload and overlaps communication with computation by maintaining algorithmic threads in each processor. To illustrate our ideas, perceptual grouping operations used in an integrated vision system for building detection are used as examples. Our experimental results show that, given 7825 line segments extracted from a 1K × 1K image, the line grouping operation can be performed in 0.294 seconds on a T3D having 64 processors. A serial implementation takes 5.696 seconds on a single processor of T3D. The implementation was performed using MPI and C and is portable.
This research was supported in part by NSF under grant CCR-9317301 and in part by DARPA under grant F49620-93-1-0620.
Supported by a scholarship from ETRI.
Chapter PDF
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Y. Chung, V. Prasanna, and C. Wang, “A Fast Asynchronous Algorithm for Linear Feature Extraction on IBM SP-2”, Proc. of Computer Architectures for Machine Perception, pp. 294–301, 1995.
A. Huertas, C. Lin, and R. Nevatia, “Detection of Buildings from Monocular Views of Aerial Scenes using Perceptual Grouping and Shadows,” Image Understanding Workshop, pp. 253–260, 1993.
H. Lu and J. Aggarwal, “Applying Perceptual Organization to the Detection of Man-Made Objects in Non-Urban Scenes”, Pattern Recognition, pp. 835–853, 1992.
V. Prasanna and C. Wang, “Scalable Parallel Implementations of Perceptual Grouping on Connection Machine CM-5”, International Conference on Pattern Recognition, pp. 229–233, 1994.
C. Wang, V. Prasanna, and Y. Lim, “Parallelization of Perceptual Grouping on Distributed Memory Machines”, Proc. of Computer Architectures for Machine Perception, pp. 323–330, 1995.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Chung, Y., Prasanna, V.K. (1996). An asynchronous parallel algorithm for symbolic grouping operations in vision. In: Bougé, L., Fraigniaud, P., Mignotte, A., Robert, Y. (eds) Euro-Par'96 Parallel Processing. Euro-Par 1996. Lecture Notes in Computer Science, vol 1124. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0024693
Download citation
DOI: https://doi.org/10.1007/BFb0024693
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-61627-6
Online ISBN: 978-3-540-70636-6
eBook Packages: Springer Book Archive