Summary
the 2014 International Symposium on Nonlinear Theory and its Applications
2014
Session Number:A2L-B
Session:
Number:A2L-B5
Highly-parallel morphological pattern spectrum processing with massive-parallel SIMD matrix architecture
Takeshi Kumaki, Yasushi Tsukada, Takeshi Ogura, Takeshi Fujino,
pp.132-135
Publication Date:2014/9/14
Online ISSN:2188-5079
DOI:10.34385/proc.46.A2L-B5
PDF download (285.6KB)
Summary:
This paper presents a highly-parallel morphological pattern spectrum processing with a massiveparallel SIMD matrix architecture (MX-1). The pattern spectrum of this method consists of a mathematical morphological-based algorithmthat can extract unique feature information from an original digital image. The morphological pattern spectrum was implemented with a MX- 1 evaluation board to verify the effectiveness of the proposed method. A well-known benchmark image was analyzed using a square structuring element from 1 to 15 scales. The obtained processing time of the MX-1 processor was up to 85% less than with four conventional processors. Therefore, the proposed parallel pattern spectrum processing method with MX-1 is effective for object extraction in mobile devices.