Abstract
A method for automated analysis of leaf margins is presented. A detailed algorithm for the decomposition of a leaf into its central part and marginal projections (teeth) is given. Cut-off projections that cannot be decomposed any further are approximated to simple geometrical figures. The set of cut-off projections is then characterized by tens of attributes, which in turn, can constitute an input for multivariate methods. In contrast to well-known shape descriptor systems (e.g., moment invariants, elliptic Fourier coefficients, chain codes) this method is highly specific, permits plausible botanical interpretation, and is computationally inexpensive. Some limitations of the approach are discussed. It seems that the technique can be utilized in computer identification systems and for purely taxonomical purposes. The paper is rounded off with cluster analysis of a small set of leaves described by marginal characters only. It is shown that leaves differing in indentation pattern but otherwise similar can be discriminated by the method. On the other hand, even considerable blade damage does not prevent the leaves of one species from being grouped together. It is also demonstrated that dendrograms based on moment invariants and elliptic Fourier coefficients do not match the predefined partition.Key words: biometrics, identification aids, image acquisition, shape descriptors.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
