Approximating morphological operators with part-based representations learned by asymmetric auto-encoders - Archive ouverte HAL Access content directly
Journal Articles Mathematical Morphology - Theory and Applications Year : 2020

Approximating morphological operators with part-based representations learned by asymmetric auto-encoders

Abstract

This paper addresses the issue of building a part-based representation of a dataset of images. More precisely, we look for a non-negative, sparse decomposition of the images on a reduced set of atoms, in order to unveil a morphological and explainable structure of the data. Additionally, we want this decomposition to be computed online for any new sample that is not part of the initial dataset. Therefore, our solution relies on a sparse, non-negative auto-encoder, where the encoder is deep (for accuracy) and the decoder shallow (for explainability). This method compares favorably to the state-of-the-art online methods on two benchmark datasets (MNIST and Fashion MNIST) and on a hyperspectral image, according to classical evaluation measures and to a new one we introduce, based on the equivariance of the representation to morphological operators.
Fichier principal
Vignette du fichier
MathMorph-D-19-00011_final.pdf (3.72 Mo) Télécharger le fichier
Origin : Publisher files allowed on an open archive
Loading...

Dates and versions

hal-02915633 , version 1 (14-08-2020)

Identifiers

Cite

Samy Blusseau, Bastien Ponchon, Santiago Velasco-Forero, Jesus Angulo, Isabelle Bloch. Approximating morphological operators with part-based representations learned by asymmetric auto-encoders. Mathematical Morphology - Theory and Applications, 2020, 4 (1), pp.64 - 86. ⟨10.1515/mathm-2020-0102⟩. ⟨hal-02915633⟩
4652 View
43 Download

Altmetric

Share

Gmail Facebook X LinkedIn More