Pixel Relationships-based Regularizer for Retinal Vessel Image Segmentation

TL;DR

This paper introduces regularizers based on graph theory and topology to incorporate pixel neighbor relationships into deep learning for retinal vessel image segmentation, improving accuracy over traditional pixel-wise loss methods.

Contribution

The study proposes novel regularizers that embed pixel neighbor relationships into the training process using graph Laplacian and Euler characteristic, enhancing segmentation performance.

Findings

Regularizers effectively capture pixel neighbor relations.

Improved segmentation accuracy over baseline models.

Regularizers reduce isolated object errors.

Abstract

The task of image segmentation is to classify each pixel in the image based on the appropriate label. Various deep learning approaches have been proposed for image segmentation that offers high accuracy and deep architecture. However, the deep learning technique uses a pixel-wise loss function for the training process. Using pixel-wise loss neglected the pixel neighbor relationships in the network learning process. The neighboring relationship of the pixels is essential information in the image. Utilizing neighboring pixel information provides an advantage over using only pixel-to-pixel information. This study presents regularizers to give the pixel neighbor relationship information to the learning process. The regularizers are constructed by the graph theory approach and topology approach: By graph theory approach, graph Laplacian is used to utilize the smoothness of segmented images based on output images and ground-truth images. By topology approach, Euler characteristic is used to identify and minimize the number of isolated objects on segmented images. Experiments show that our scheme successfully captures pixel neighbor relations and improves the performance of the convolutional neural network better than the baseline without a regularization term.