Abstract
BackgroundGeometric morphometric methods of capturing information about curves or outlines of organismal structures may be used in conjunction with canonical variates analysis (CVA) to assign specimens to groups or populations based on their shapes. This methodological paper examines approaches to optimizing the classification of specimens based on their outlines. This study examines the performance of four approaches to the mathematical representation of outlines and two different approaches to curve measurement as applied to a collection of feather outlines. A new approach to the dimension reduction necessary to carry out a CVA on this type of outline data with modest sample sizes is also presented, and its performance is compared to two other approaches to dimension reduction.ResultsTwo semi-landmark-based methods, bending energy alignment and perpendicular projection, are shown to produce roughly equal rates of classification, as do elliptical Fourier methods and the extended eigenshape method of outline measurement. Rates of classification were not highly dependent on the number of points used to represent a curve or the manner in which those points were acquired. The new approach to dimensionality reduction, which utilizes a variable number of principal component (PC) axes, produced higher cross-validation assignment rates than either the standard approach of using a fixed number of PC axes or a partial least squares method.ConclusionClassification of specimens based on feather shape was not highly dependent of the details of the method used to capture shape information. The choice of dimensionality reduction approach was more of a factor, and the cross validation rate of assignment may be optimized using the variable number of PC axes method presented herein.
Highlights
Geometric morphometric methods of capturing information about curves or outlines of organismal structures may be used in conjunction with canonical variates analysis (CVA) to assign specimens to groups or populations based on their shapes
Since the automated curve tracing required manually tracing the feather outline for our data set to get reliable detection of the outline edge, due to the irregular edges of the feathers, and did not appear to offer higher repeatability (Table 1), we did not attempt to carry out automated curve tracing on the remaining specimens
(page number not for citation purposes) http://www.frontiersinzoology.com/content/3/1/15 acteristics of CVA indicate the promise offered by semilandmark methods of integrating information about outlines or curves with landmark points
Summary
Geometric morphometric methods of capturing information about curves or outlines of organismal structures may be used in conjunction with canonical variates analysis (CVA) to assign specimens to groups or populations based on their shapes. This methodological paper examines approaches to optimizing the classification of specimens based on their outlines. The use of multivariate statistical methods ( canonical variates analysis, CVA, a multiple group form of discriminant analysis) to classify specimens into groups requires that the curves or outlines on the specimens be represented by a limited number of measured variables. Classification of specimens based on outlines poses a challenge, in that accurate representation of a curve requires many measurements accurately, but this increase in parameters dramatically increases the sample sizes necessary to carry out the CVA
Sign-in/Register to view highlights & summary 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.
