Data Augmentation in Training CNNs: Injecting Noise to Images

Murtaza Eren Akbiyik

Keywords: deep learning, data augmentation, convolutional neural networks, noise, image processing, SSIM

TL;DR: Ideal methodology to inject noise to input data during CNN training

Abstract: Noise injection is a fundamental tool for data augmentation, and yet there is no widely accepted procedure to incorporate it with learning frameworks. This study analyzes the effects of adding or applying different noise models of varying magnitudes to Convolutional Neural Network (CNN) architectures. Noise models that are distributed with different density functions are given common magnitude levels via Structural Similarity (SSIM) metric in order to create an appropriate ground for comparison. The basic results are conforming with the most of the common notions in machine learning, and also introduces some novel heuristics and recommendations on noise injection. The new approaches will provide better understanding on optimal learning procedures for image classification.

Code: https://drive.google.com/open?id=1GwQFo2QtW_O6AebR-ZGUYqlv6ItVx2_2

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2307.06855/code)