Pré-Publication, Document De Travail Année : 2025

Learning Morphological Representations of Image Transformations: Influence of Initialization and Layer Differentiability

Résumé

As a combination of two successful paradigms in image processing, namely Mathematical Morphology and Deep Learning, Morphological Networks seem very promising for image analysis. However, their practical applications remain constrained due to possible limitations of expressivity and difficulties related to gradient descent based optimization algorithms. In this paper, we focus on neural architectures inspired by the morphological representation theory, which supports their expressivity. We investigate their optimization difficulties and find out that, rather than the non-differentiability of morphological layers, the sparsity of their gradient (or approximated gradient) is what limits them the most, and requires an appropriate initialization. Furthermore, we propose a method to reduce the number of model parameters after training, which produces a minimal equivalent operator.

Fichier principal
Vignette du fichier
DGMM_2025_paper_22.pdf (1.03 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-05112110 , version 1 (13-06-2025)
hal-05112110 , version 2 (04-11-2025)

Licence

Identifiants

  • HAL Id : hal-05112110 , version 1

Citer

Mihaela Dimitrova, Samy Blusseau, Santiago Velasco-Forero. Learning Morphological Representations of Image Transformations: Influence of Initialization and Layer Differentiability. 2025. ⟨hal-05112110v1⟩
170 Consultations
405 Téléchargements

Partager

  • More