Domain-decomposed image classification algorithms using linear discriminant analysis and convolutional neural networks

TL;DR

This paper compares domain-decomposed CNN models with transfer learning and introduces a novel localized LDA approach, both showing improved accuracy and training efficiency in image classification tasks.

Contribution

It presents two new domain-decomposed CNN models with transfer learning and a novel localized LDA strategy, enhancing accuracy and training speed.

Findings

Decomposed CNN models outperform global CNN without transfer learning.

Localized LDA achieves higher accuracy than global LDA.

Both methods improve training efficiency.

Abstract

In many modern computer application problems, the classification of image data plays an important role. Among many different supervised machine learning models, convolutional neural networks (CNNs) and linear discriminant analysis (LDA) as well as sophisticated variants thereof are popular techniques. In this work, two different domain decomposed CNN models are experimentally compared for different image classification problems. Both models are loosely inspired by domain decomposition methods and in addition, combined with a transfer learning strategy. The resulting models show improved classification accuracies compared to the corresponding, composed global CNN model without transfer learning and besides, also help to speed up the training process. Moreover, a novel decomposed LDA strategy is proposed which also relies on a localization approach and which is combined with a small neural network model. In comparison with a global LDA applied to the entire input data, the presented decomposed LDA approach shows increased classification accuracies for the considered test problems.