| | |
Summary: Pipelined Implementation of the Hough
Transform in a Pyramid Multiprocessor
M. Atiquzzaman
Dept. of Computer Science
La Trobe University, Melbourne 3083, Australia.
Email: atiq@LATCS1.lat.oz.au
Abstract
The Hough transform is used to detect patterns in images. Efficient algorithms
and their implementations in multiprocessor systems have been proposed to over
come the computational requirements. An efficient pipelined implementation of the
transform in a pyramid machine has been proposed. The proposed implementation
considers task scheduling to ensure uniform computational load among the proces
sor elements in the pyramid. The Multiresolution Hough transform (MHT) has
been selected for implementation in the pyramid machine.
Index terms: Pyramid architecture, Hough transform, Image processing, Multiresolu
tion analysis, Pipelined processing.
1 Introduction
The Hough transform [1] is a powerful technique to detect patterns in images. The
transform maps patterns from the image plane to a parameter space. The advantages of
the transform are its robustness to noise in the image and discontinuities in the pattern
|
