Improvement of image transform calculation based on a weighted primitive

  • Authors:
  • María Teresa Signes Pont;Juan Manuel García Chamizo;Higinio Mora Mora;Gregorio de Miguel Casado

  • Affiliations:
  • Departamento de Tecnología Informática y Computación, University of Alicante, San Vicente del Raspeig, Alicante, Spain;Departamento de Tecnología Informática y Computación, University of Alicante, San Vicente del Raspeig, Alicante, Spain;Departamento de Tecnología Informática y Computación, University of Alicante, San Vicente del Raspeig, Alicante, Spain;Departamento de Tecnología Informática y Computación, University of Alicante, San Vicente del Raspeig, Alicante, Spain

  • Venue:
  • ICIAR'06 Proceedings of the Third international conference on Image Analysis and Recognition - Volume Part I
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents a method to improve the calculation of functions that demand a great amount of computing resources. The fundamentals argue for an increase of the computing power of the primitive level in order to decrease the number of computing levels required to carry out calculations. A weighted primitive substitutes the usual primitives sum and multiplication and calculates the function values by successive iterations. The parametric architecture associated to the weighted primitive is particularly suitable in the case of combined trigonometric functions sine and cosine involved in the calculation of image transforms. The Hough Transform (HT) and the Fourier Transform (FT) are analyzed under this scope, obtaining a good performance and trade-off between speed and area requirements when comparing with other well-known proposals.