Title :
Neural evaluation of texture synthesis results
Author :
Rouis, Kais ; Sayadi, Mounir ; Fnaiech, Farhat
Author_Institution :
Higher Sch. of Sci. & Tech., Univ. of Tunis, Tunis, Tunisia
Abstract :
This paper introduces a new idea of using the Self-Organizing Map to compare the similarity between the synthesized textures and the original sample in accordance with human visual system. The approaches of texture synthesis do not provide convincing results for all types of textures, since it is difficult to produce a general definition of this term. So that, we propose to evaluate the quality of results provided by different approaches, based on features of comparison to capture information present in texture image.
Keywords :
image matching; image texture; self-organising feature maps; human visual system; image texture synthesis; neural evaluation; original textures; self-organizing map; synthesized textures; texture similarity; Feature extraction; Image color analysis; Indexes; Neurons; Vectors; Visualization; Wavelet transforms;
Conference_Titel :
Electrical Engineering and Software Applications (ICEESA), 2013 International Conference on
Conference_Location :
Hammamet
Print_ISBN :
978-1-4673-6302-0
DOI :
10.1109/ICEESA.2013.6578466