Chapter 1: Invited Addresses and Tutorials on Signals, Coding,
  Systems and Intelligent Techniques

Title: Image segmentation using pulse coupled neural networks

Author(s): Panos Liatsis, Bo Li, Tomislav Kos, Ida Srdic

Address: Information and Biomedical Engineering Centre, School of Engineering and Mathematical Sciences City University, London, United Kingdom | Information and Biomedical Engineering Centre, School of Engineering and Mathematical Sciences City University, London, United Kingdom | Faculty of Electrical Engineering and Computing, University of Zagreb, Zagreb, Croatia | Croatian Telecom, Zagreb, Croatia

Reference: 12th International Workshop on Systems, Signals and Image Processing pp. 129 - 133

Abstract/Summary: This manuscript deals with the application of Seeded Region Growing (SRG) and Pulse Coupled Neural Networks (PCNN) in image segmentation. SRG selects the initial seed points for the initialisation of the algorithm. During this process, the neighbours of the seed pixels are compared to certain similarity criteria and are appropriately merged to regions. PCNN have recently become popular in image processing applications. They have a dedicated architecture consisting of feeding/linking fields and spike generator and do not require training. The performance of the two techniques will be thoroughly investigated with applications to cDNA microarray images.

Order a copy of this article