article · 2007
A Monolithic Mixed-Mode Implementation of Sum-of-Product Arrays for Performing Binary Morphological Image Processing
Ronald Spencer And
Cited by 0Abstract and details only
Abstract
This paper presents a monolithic mixed-mode implementation of sum-of-product arrays that are capable of performing morphological image processing in silicon. The digital equations of the two basic morphological operations, erosion and dilation, are rewritten in a local, mixed-mode form to facilitate hardware implementation and results are shown for a 3x3 selective element array fabricated in silicon. I. INTRODUCTION Morphology is the study of structure in images [1] and [2]. It is based on the concept of transforming images by probing them with smaller images known as selective elements to determine intersections and unions. Two very significant advantages of morphological processing are the ability to discover fundamental structures in images and the localized simplicity of the operations. As a result, morphological processing can be easily implemented in hardware to assist with such tasks as image analysis and video coding [3]. Although morphological operations can be implemented in...
Cite this paper
And, R. S. (2007). A Monolithic Mixed-Mode Implementation of Sum-of-Product Arrays for Performing Binary Morphological Image Processing.
And, Ronald Spencer. “A Monolithic Mixed-Mode Implementation of Sum-of-Product Arrays for Performing Binary Morphological Image Processing.” 2007.
And, Ronald Spencer. 2007. “A Monolithic Mixed-Mode Implementation of Sum-of-Product Arrays for Performing Binary Morphological Image Processing.”
And, R.S. (2007) ‘A Monolithic Mixed-Mode Implementation of Sum-of-Product Arrays for Performing Binary Morphological Image Processing’.
@article{and2007monolithic,
title = {{A Monolithic Mixed-Mode Implementation of Sum-of-Product Arrays for Performing Binary Morphological Image Processing}},
author = {And, Ronald Spencer},
year = {2007}
}Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.