Binary morphological neural network - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2022

Binary morphological neural network

Théodore Aouad
Hugues Talbot

Abstract

In the last ten years, Convolutional Neural Networks (CNNs) have formed the basis of deep-learning architectures for most computer vision tasks. However, they are not necessarily optimal. For example, mathematical morphology is known to be better suited to deal with binary images. In this work, we create a morphological neural network that handles binary inputs and outputs. We propose their construction inspired by CNNs to formulate layers adapted to such images by replacing convolutions with erosions and dilations. We give explainable theoretical results on whether or not the resulting learned networks are indeed morphological operators. We present promising experimental results designed to learn basic binary operators, and we have made our code publicly available online.
Fichier principal
Vignette du fichier
Bimonn_ICIP_2022_Preprint_for_Arxiv.pdf (434.43 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03617249 , version 1 (23-03-2022)

Identifiers

  • HAL Id : hal-03617249 , version 1

Cite

Théodore Aouad, Hugues Talbot. Binary morphological neural network. IEEE International Conference on Image Processing (ICIP), Oct 2022, Bordeaux, France. pp.3276-3280. ⟨hal-03617249⟩
116 View
147 Download

Share

Gmail Mastodon Facebook X LinkedIn More