An MLP Baseline for Handwriting Recognition Using Planar Curvature and Gradient Orientation

TL;DR

This paper demonstrates that a multilayer perceptron using handcrafted geometric features like curvature and gradient orientation can achieve high accuracy in handwritten character recognition, offering an interpretable alternative to CNNs.

Contribution

It introduces a curvature and gradient orientation-based feature set for MLPs, showing competitive performance without deep learning architectures.

Findings

97% accuracy on MNIST digits

89% accuracy on EMNIST letters

Handcrafted geometric features are highly discriminative

Abstract

This study investigates whether second-order geometric cues - planar curvature magnitude, curvature sign, and gradient orientation - are sufficient on their own to drive a multilayer perceptron (MLP) classifier for handwritten character recognition (HCR), offering an alternative to convolutional neural networks (CNNs). Using these three handcrafted feature maps as inputs, our curvature-orientation MLP achieves 97 percent accuracy on MNIST digits and 89 percent on EMNIST letters. These results underscore the discriminative power of curvature-based representations for handwritten character images and demonstrate that the advantages of deep learning can be realized even with interpretable, hand-engineered features.