Rotational Augmentation Techniques: A New Perspective on Ensemble Learning for Image Classification

Abstract

The popularity of data augmentation techniques in machine learning has increased in recent years, as they enable the creation of new samples from existing datasets. Rotational augmentation, in particular, has shown great promise by revolving images and utilising them as additional data points for training. The research in this study aimed to evaluate the effectiveness of rotational augmentation techniques and different voting systems in improving image classification accuracy. To accomplish this, several image datasets were evaluated using various augmentation methods, which were employed to generate testing sets. Subsequently, voting systems were used to determine the most reliable outcome for each original data. The findings of this study suggest that rotational augmentation techniques can significantly enhance the accuracy of classification models. Additionally, the selection of a voting scheme can considerably impact the model's performance. Overall, the study found that using an ensemble-based voting system produced more accurate results than simple voting.