Image Classification Using Singular Value Decomposition and Optimization

TL;DR

This paper explores using Singular Value Decomposition combined with optimization techniques for classifying cat and dog breeds based on fur color, achieving moderate accuracy and highlighting the potential and limitations of low-rank image features.

Contribution

It introduces a novel approach combining SVD and SQP for breed classification, demonstrating the effectiveness of low-rank approximations in this context.

Findings

Achieved 69% accuracy at rank 10 using Frobenius norm.

Partial validation of fur color as a dominant feature in low-rank approximations.

Highlights the trade-off between simplicity and accuracy in resource-limited settings.

Abstract

This study investigates the applicability of Singular Value Decomposition for the image classification of specific breeds of cats and dogs using fur color as the primary identifying feature. Sequential Quadratic Programming (SQP) is employed to construct optimally weighted templates. The proposed method achieves 69% accuracy using the Frobenius norm at rank 10. The results partially validate the assumption that dominant features, such as fur color, can be effectively captured through low-rank approximations. However, the accuracy suggests that additional features or methods may be required for more robust classification, highlighting the trade-off between simplicity and performance in resource-constrained environments.