Title: A method for identification of irregular objects under noisy conditions

Authors: Sándor Molnár; Péter Klinkó; István Szabó

Addresses: Department of Informatics, Faculty of Mechanical Engineering, Szent István University, Páter K. u.1, H-2103 Gödöllo, Hungary. ' Institute of Mechanics and Machinery, Faculty of Mechanical Engineering, Szent István University, Páter K. u.1, H-2103 Gödöllo, Hungary. ' Institute of Mechanics and Machinery, Faculty of Mechanical Engineering, Szent István University, Páter K. u.1, H-2103 Gödöllo, Hungary

Abstract: In the present paper, we propose a method of detecting amorphously shaped objects in real-time ultrasonic images when high level of noise is present. In general, noisy imaging environment can lead to faulty object segmentation. Shape recognition algorithms can give ambiguous results even for detecting objects with permanent shapes. From prior knowledge, indexed object templates are introduced that are aimed to incorporate the most relevant object.

Keywords: shape detection; pattern recognition; image segmentation; object identification; irregular objects; ultrasonic images; noise levels; noisy imaging; shape recognition.

DOI: 10.1504/IJSSE.2012.046563

International Journal of System of Systems Engineering, 2012 Vol.3 No.1, pp.24 - 32

Published online: 16 Aug 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article