Fast multiplication algorithm for square sparse matrices. Application to
images processing
Ștefan-Daniel Achirei\,^1, Antonio Lasanta\,^2, Laiachi El Kaoutit\,^3 , and Carlos Rodriguez Dominguez\,^4
\,^1\,“Gheorghe Asachi” Technical University of Lasi.
\,^2\,Universidad de Granada, Departamento de Álgebra. Facultad de Educación, Econonía y Tecnología de Ceuta.
Cortadura del Valle, s/n. E-51001 Ceuta, Spain. Instituto Carlos I de Física Teórica y Computacional,
Universidad de Granada, 18071 Granada, Spain.
\,^3\,Universidad de Granada, Departamento de Álgebra and IMAG. Facultad de Ciencias s/n. E-18071 Granada, Spain.
\,^4\,Universidad de Granada, Departamento de Lenguajes y Sistema Informáticos.
Facultad de Educación, Econonía y Tecnología de Ceuta. Cortadura del Valle, s/n. E-51001 Ceuta, Spain.
Pages 197-218 | Received 14 November 2023, Accepted 14 January 2024, Published 30 June 2024
Abstract
We provide a fast multiplication algorithm for a certain class of square sparse matrices that are commonly used in image processing. In order to compare our approach with the standard algorithms, we will use two working sets: SuitSparse Matrix Collection and Anonymous MRI Brain Scan Images Database. The evaluations show that our algorithm has up to 75 times better time-efficiency and an improvement between 21% and 96% of memory-efficiency.
Keywords: Digital Images, Image Processing, Gray-scale, Sparse Matrices, Software Algorithms, Iterative Algorithms.
MSC numbers: 65Y20, 68W01, 68W35.
Downloads: Full-text PDF