An approach based on class activation maps for investigating the effects of data augmentation on neural networks for image classification

TL;DR

This paper introduces a methodology using class activation maps to quantitatively analyze how various data augmentation strategies influence the learned patterns and robustness of convolutional neural networks in image classification tasks.

Contribution

It proposes a novel approach and metrics for analyzing the effects of data augmentation on neural network models using class activation maps.

Findings

Data augmentation impacts can be quantitatively measured.

Different augmentation strategies produce distinct impact profiles.

The methodology helps identify how models focus on image regions with augmentation.

Abstract

Neural networks have become increasingly popular in the last few years as an effective tool for the task of image classification due to the impressive performance they have achieved on this task. In image classification tasks, it is common to use data augmentation strategies to increase the robustness of trained networks to changes in the input images and to avoid overfitting. Although data augmentation is a widely adopted technique, the literature lacks a body of research analyzing the effects data augmentation methods have on the patterns learned by neural network models working on complex datasets. The primary objective of this work is to propose a methodology and set of metrics that may allow a quantitative approach to analyzing the effects of data augmentation in convolutional networks applied to image classification. An important tool used in the proposed approach lies in the concept of class activation maps for said models, which allow us to identify and measure the importance these models assign to each individual pixel in an image when executing the classification task. From these maps, we may then extract metrics over the similarities and differences between maps generated by these models trained on a given dataset with different data augmentation strategies. Experiments made using this methodology suggest that the effects of these data augmentation techniques not only can be analyzed in this way but also allow us to identify different impact profiles over the trained models.