A Learning Framework for Morphological Operators using Counter-Harmonic Mean

TL;DR

This paper introduces a new learning framework that combines morphological operators with neural networks, enabling automatic learning of structuring elements and operator compositions, validated through experiments including real-world steel industry applications.

Contribution

It proposes a novel framework that integrates morphological operations with neural network learning, allowing automatic adaptation of operators and structuring elements from data.

Findings

Successfully learns morphological operators and structuring elements

Scales efficiently to large datasets and online learning

Demonstrates real-world application in steel industry

Abstract

We present a novel framework for learning morphological operators using counter-harmonic mean. It combines concepts from morphology and convolutional neural networks. A thorough experimental validation analyzes basic morphological operators dilation and erosion, opening and closing, as well as the much more complex top-hat transform, for which we report a real-world application from the steel industry. Using online learning and stochastic gradient descent, our system learns both the structuring element and the composition of operators. It scales well to large datasets and online settings.