- Spiral Architecture is a relatively new and powerful approach to general purpose machine vision system. It contains very useful geometric and algebraic properties. Two algebraic operations, Spiral Addition and Spiral Multiplication, have been defined on it. This paper presents a way to segment the object(s) in an image uniformly by Spiral Multiplication. Namely, a number of analogous small copies of the original object(s) are made during segmentation. An algorithm is also developed in this paper to compute the scaling factor or the number of the small copies, so image segmentation can be done flexibly and quantitatively according to the specific application. The research results are very beneficial to image segmentation in parallel image processing and distributed image processing.
(from crpit.com)
(local if available)
