Aggregated pairwise classification of elastic planar shapes

DOI10.1214/21-AOAS1452zbMATH Open1478.62161arXiv1901.07593OpenAlexW3182441629MaRDI QIDQ2245135FDOQ2245135

Authors: Min Ho Cho, Sebastian Kurtek, Steven N. MacEachern 

Publication date: 15 November 2021

Published in: The Annals of Applied Statistics (Search for Journal in Brave)

Abstract: The classification of shapes is of great interest in diverse areas ranging from medical imaging to computer vision and beyond. While many statistical frameworks have been developed for the classification problem, most are strongly tied to early formulations of the problem - with an object to be classified described as a vector in a relatively low-dimensional Euclidean space. Statistical shape data have two main properties that suggest a need for a novel approach: (i) shapes are inherently infinite dimensional with strong dependence among the positions of nearby points, and (ii) shape space is not Euclidean, but is fundamentally curved. To accommodate these features of the data, we work with the square-root velocity function of the curves to provide a useful formal description of the shape, pass to tangent spaces of the manifold of shapes at different projection points which effectively separate shapes for pairwise classification in the training data, and use principal components within these tangent spaces to reduce dimensionality. We illustrate the impact of the projection point and choice of subspace on the misclassification rate with a novel method of combining pairwise classifiers.

Full work available at URL: https://arxiv.org/abs/1901.07593

Recommendations

• Random planar shapes and their statistical recognition
• Morphological analysis of cells by means of an elastic metric in the shape space
• On shape of plane elastic curves
• Elastic statistical shape analysis of biological structures with case studies: a tutorial
• Riemannian Structures on Shape Spaces: A Framework for Statistical Inferences

zbMATH Keywords

dimension reductionregistrationLDApairwise classificationQDAprojection point

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Applications of statistics in engineering and industry; control charts (62P30) Image analysis in multivariate analysis (62H35)

Cites Work

• Functional and shape data analysis
• Shape Manifolds, Procrustean Metrics, and Complex Projective Spaces
• Computable Elastic Distances Between Shapes
• Statistical modeling of curves using shapes and related features
• Statistical shape analysis. With applications in R
• The Intrinsic Bayes Factor for Model Selection and Prediction
• Bayesian synthesis: combining subjective analyses, with an application to ozone data
• Nested sphere statistics of skeletal models
• Medial representations. Mathematics, algorithms and applications
• On shape of plane elastic curves
• Large deformation diffeomorphic metric curve mapping
• Fourier Descriptors for Plane Closed Curves
• Computational anatomy: an emerging discipline
• How to conjugate C\(^1\)-close group actions
• Size and shape analysis of landmark data
• A fast level set based algorithm for topology-independent shape modeling.

Cited In (3)

• Deformation analysis for shape based classification
• Logistic regression models for elastic shape of curves based on tangent representations
• Title not available (Why is that?)

Uses Software

• Flavia

This page was built for publication: Aggregated pairwise classification of elastic planar shapes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2245135)