Fringe Pattern Images as Visual Representation-Based Approach in Structured Data Classification

Conventional models often fall short in supporting decision-making from structured tabular data, as they struggle to capture complex feature interactions. At the same time, convolutional neural networks (CNNs), highly effective in computer vision, remain underutilized in this context. We propose a method to bridge this gap by transforming tabular records into fringe pattern images using Gaussian surface modeling. This visual encoding enables CNN-based classification and is evaluated across architectures such as ResNet, DenseNet, and GoogleNet. Our approach outperforms baseline models, highlighting the potential of visual representations to enhance structured data analysis across diverse applications.