Data augmentation of a handwritten character dataset for a Convolutional Neural Network and integration into a Bayesian Linear Framework

Abstract

Convolutional neural networks are often used for image recognition. They have, for example, achieved the lowest error rate (0.23 percent) for the MNIST database up until now [6]. The use of receptive elds also makes them similar to how the human visual cortex works. The task at hand is to use convolutional neural networks to nd optimizations for brainreading research [21]. The goal is to nd these optimizations through using di erent preprocessing methods on the handwritten character dataset [27] and testing which method results in the highest classi cation accuracy for the convolutional neural networks. An optimization is achieved by using the most e cient data augmentation methods from the convolutional neural networks to preprocess the prior of the Bayesian linear framework.