You can view the full text of this article for free using the link below.

Title: Generic filtering and removing artefacts from document images using unsupervised PSO optimisation

Authors: Aicha Eutamene; Djamel Gaceb; Hacene Belhadef

Addresses: Department of Fundamental Computer Science and Its Applications, NTIC Faculty, MISC Laboratory, University of Constantine 2, Ali Mendjeli City, Constantine 25017, Algeria ' Department of Computer Science, Faculty of Sciences, LIMOSE Laboratory, UMBB University, Boumerdes 35000, Algeria ' Department of Fundamental Computer Science and Its Applications, NTIC Faculty, MISC Laboratory, University of Constantine 2, Ali Mendjeli City, Constantine 25017, Algeria

Abstract: The advancements in the field of analysis and optical recognition of document images have accelerated recently due to the many emerging applications which are not only challenging but also computationally more demanding, such as mail and document sorting, automatic classification of documents, handwriting and script recognition, etc. In this paper, our contribution focuses on preprocessing of these applications: smoothing and filtering of degraded document images using a new adaptive mean shift algorithm based on the integral image. The great difficulty of parameter setting of this approach requires solving of complex optimisation problems using metaheuristic algorithms. Our goal is to demonstrate the contribution of the particle swarm optimisation (PSO) method to improve the quality and the parameter setting of the developed preprocessing approach. We tested and compared two types of objective functions (supervised and unsupervised) and demonstrate the effectiveness of the optimisation in an unsupervised context.

Keywords: degraded documents; image pre-processing; image smoothing; adaptive mean shift; metaheuristics; OCR systems; optical character recognition; PSO; particle swarm optimisation; generic filtering; artefact removal; degraded images; swarm intelligence.

DOI: 10.1504/IJMHEUR.2017.083097

International Journal of Metaheuristics, 2017 Vol.6 No.1/2, pp.55 - 84

Received: 09 Mar 2016
Accepted: 14 Nov 2016

Published online: 20 Mar 2017 *

Full-text access for editors Access for subscribers Free access Comment on this article