Please use this identifier to cite or link to this item:
Title: COSFIRE : a trainable features approach to pattern recognition
Authors: Azzopardi, George
Petkov, Nicolai
Keywords: Pattern recognition systems
Computer vision
Computer graphics
Data structures (Computer science)
Artificial intelligence
Optical pattern recognition
Issue Date: 2013
Publisher: Radboud University Nijmegen
Citation: Azzopardi, G., & Petkov, N. (2013). COSFIRE : a trainable features approach to pattern recognition. BENELEARN 2013, Nijmegen.
Abstract: In a recent work (Azzopardi & Petkov, 2013), we proposed a trainable features approach to visual pattern recognition. It is called COSFIRE, which stands for Combination of Shifted Filter Responses. A COSFIRE operator is automatically configured by a specified pattern of interest, referred to as a prototype, and is then able to detect the same and similar patterns in other images. The configuration comprises the determination of the orientations of dominant contour parts and their mutual spatial arrangement.
Appears in Collections:Scholarly Works - FacICTAI

Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.