PIPELINED IMPLEMENTATION OF THE MULTIRESOLUTION HOUGH TRANSFORM IN A PYRAMID MULTIPROCESSOR

The Hough transform is used to detect patterns in images. Its advantag e is the ability to detect discontinuous patterns in noisy images. The drawback is its demand for a large amount of computing power. Efficie nt algorithms and their implementations in multiprocessor systems have been proposed to overcome the computational requirements. The ability to efficiently may two-dimensional (2D) images into 2D meshes makes t he pyramid computer a very attractive candidate for implementation of image processing algorithms. Several implementations of the Hough tran sform in pyramid machines can be found in the literature. The implemen tations are geared towards the analysis of a single image. Moreover, t he implementations do not consider the balancing of the computational load uniformly among the processor elements in the pyramid. The object ive of this paper is to propose an efficient pipelined implementation of the multiresolution Hough transform in pyramid machine. The propose d implementation considers task scheduling to ensure balanced computat ional load among the processor elements in the pyramid. It is shown th at the proposed pipelined implementation is suitable for efficiency an alyzing a continuous sequence of images, as found in a typical industr ial inspection system. The multiresolution images and accumulator arra ys used in the multiresolution Hough transform (MHT) mirrors the varia ble sized 2D processor arrays in the pyramid layers. The MHT has there fore, been used for implementation in the pyramid machine.