This chapter presents a spatial color algorithm called Termite Retinex and the problem of filtering locality in this family of algorithms for image enhancement, inspired by the human vision system. The algorithm we present is a recent implementation of Retinex with a colony of agents, which uses swarm intelligence to explore the image, determining in this way the locality of its filtering. This results in an unsupervised detail enhancement, dynamic range stretching, color correction, and high dynamic range tone rendering. In the chapter we describe the characteristic of glocality ( glocal=global+local ) of image exploration, and after a description of the Retinex spatial color algorithm family, we present the Termite approach and discuss results.

Ant colony for locality foraging in image enhancement / G. Simone, D. Gadia, I. Farup, A. Rizzi (STUDIES IN COMPUTATIONAL INTELLIGENCE). - In: Multi-objective swarm intelligence : theoretical advances and applications / [a cura di] D. Satchidananda, A.K. Jagadev, P. Mrutyunjaya. - New York : Springer, 2015. - ISBN 9783662463086. - pp. 123-142 [10.1007/978-3-662-46309-3_5]

Ant colony for locality foraging in image enhancement

D. Gadia;A. Rizzi
2015

Abstract

This chapter presents a spatial color algorithm called Termite Retinex and the problem of filtering locality in this family of algorithms for image enhancement, inspired by the human vision system. The algorithm we present is a recent implementation of Retinex with a colony of agents, which uses swarm intelligence to explore the image, determining in this way the locality of its filtering. This results in an unsupervised detail enhancement, dynamic range stretching, color correction, and high dynamic range tone rendering. In the chapter we describe the characteristic of glocality ( glocal=global+local ) of image exploration, and after a description of the Retinex spatial color algorithm family, we present the Termite approach and discuss results.
retinex; spatial color algorithm; artificial termites; image enhancement
Settore INF/01 - Informatica
2015
Book Part (author)
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2434/266292
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 4
  • ???jsp.display-item.citation.isi??? 4
social impact