Reservoir Computing Approach for Gray Images Segmentation

TL;DR

This paper introduces a reservoir computing method utilizing echo state networks to extract features from grayscale images, enhancing segmentation accuracy through equilibrium states and intrinsic plasticity tuning, demonstrated on the Lena benchmark.

Contribution

It presents a novel reservoir computing approach that leverages equilibrium states and intrinsic plasticity for improved grayscale image segmentation.

Findings

Enhanced segmentation quality on Lena benchmark

Reservoir equilibrium states reveal hidden image features

Intrinsic plasticity tuning improves segmentation accuracy

Abstract

The paper proposes a novel approach for gray scale images segmentation. It is based on multiple features extraction from single feature per image pixel, namely its intensity value, using Echo state network. The newly extracted features - reservoir equilibrium states - reveal hidden image characteristics that improve its segmentation via a clustering algorithm. Moreover, it was demonstrated that the intrinsic plasticity tuning of reservoir fits its equilibrium states to the original image intensity distribution thus allowing for its better segmentation. The proposed approach is tested on the benchmark image Lena.