Deep Belief Networks for Image Denoising

TL;DR

This paper introduces a novel image denoising method using Deep Belief Networks that learns to distinguish noise features from image features, achieving significant noise reduction on the MNIST dataset.

Contribution

The paper presents a new approach leveraging DBNs to identify and deactivate noise-related nodes, improving image denoising performance.

Findings

Achieved 65.9% reduction in mean square error on MNIST with added noise.

Demonstrated DBNs' effectiveness in feature-based noise separation.

Validated the method's potential for image denoising tasks.

Abstract

Deep Belief Networks which are hierarchical generative models are effective tools for feature representation and extraction. Furthermore, DBNs can be used in numerous aspects of Machine Learning such as image denoising. In this paper, we propose a novel method for image denoising which relies on the DBNs' ability in feature representation. This work is based upon learning of the noise behavior. Generally, features which are extracted using DBNs are presented as the values of the last layer nodes. We train a DBN a way that the network totally distinguishes between nodes presenting noise and nodes presenting image content in the last later of DBN, i.e. the nodes in the last layer of trained DBN are divided into two distinct groups of nodes. After detecting the nodes which are presenting the noise, we are able to make the noise nodes inactive and reconstruct a noiseless image. In section 4 we explore the results of applying this method on the MNIST dataset of handwritten digits which is corrupted with additive white Gaussian noise (AWGN). A reduction of 65.9% in average mean square error (MSE) was achieved when the proposed method was used for the reconstruction of the noisy images.