Deep Convolutional Neural Networks Meet Variational Shape Compactness Priors for Image Segmentation

TL;DR

This paper introduces novel primal-dual algorithms for shape-compactness constrained image segmentation, integrating them into deep neural networks to improve efficiency and accuracy, especially on noisy datasets.

Contribution

The paper proposes two new primal-dual algorithms, PD-TD and PD-STD, for shape-compactness constrained segmentation, and demonstrates their integration into deep learning models for improved performance.

Findings

Outperformed state-of-the-art algorithms in efficiency and effectiveness.

Achieved higher IoU, dice, and compactness metrics on noisy datasets.

Significantly improved IoU by 20% on highly noisy images.

Abstract

Shape compactness is a key geometrical property to describe interesting regions in many image segmentation tasks. In this paper, we propose two novel algorithms to solve the introduced image segmentation problem that incorporates a shape-compactness prior. Existing algorithms for such a problem often suffer from computational inefficiency, difficulty in reaching a local minimum, and the need to fine-tune the hyperparameters. To address these issues, we propose a novel optimization model along with its equivalent primal-dual model and introduce a new optimization algorithm based on primal-dual threshold dynamics (PD-TD). Additionally, we relax the solution constraint and propose another novel primal-dual soft threshold-dynamics algorithm (PD-STD) to achieve superior performance. Based on the variational explanation of the sigmoid layer, the proposed PD-STD algorithm can be integrated into Deep Neural Networks (DNNs) to enforce compact regions as image segmentation results. Compared to existing deep learning methods, extensive experiments demonstrated that the proposed algorithms outperformed state-of-the-art algorithms in numerical efficiency and effectiveness, especially while applying to the popular networks of DeepLabV3 and IrisParseNet with higher IoU, dice, and compactness metrics on noisy Iris datasets. In particular, the proposed algorithms significantly improve IoU by 20% training on a highly noisy image dataset.