Abstract
This paper presents a new measure of symmetry for bifurcating structures, which relies not only on topology and ordering, but also on quantitative properties (e.g. length of branches). This measure is based on a specific biological mechanism and on the concept of minimum energy. The effectiveness of the approach is demonstrated in a classification test where leaves taken from plants growing under different stress conditions are classified. Results show that the proposed measure improves classification performance compared to classification based on other leading measures.
Original language | English |
---|---|
Pages (from-to) | 2237-2250 |
Number of pages | 14 |
Journal | Pattern Recognition |
Volume | 40 |
Issue number | 8 |
DOIs | |
State | Published - Aug 2007 |
Keywords
- Bifurcating structures
- CSM
- Continuous symmetry
- Graphs
- Leaf veins
- Shape characteristics
- Symmetry
ASJC Scopus subject areas
- Software
- Signal Processing
- Computer Vision and Pattern Recognition
- Artificial Intelligence