Research Article
An MLP Baseline for Handwriting Recognition Using Planar Curvature and Gradient Orientation
|
International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
|
| Volume 187 - Issue 47 |
| Published: October 2025 |
| Authors: Azam Nouri |
10.5120/ijca2025925791
|
Azam Nouri . An MLP Baseline for Handwriting Recognition Using Planar Curvature and Gradient Orientation. International Journal of Computer Applications. 187, 47 (October 2025), 1-5. DOI=10.5120/ijca2025925791
@article{ 10.5120/ijca2025925791,
author = { Azam Nouri },
title = { An MLP Baseline for Handwriting Recognition Using Planar Curvature and Gradient Orientation },
journal = { International Journal of Computer Applications },
year = { 2025 },
volume = { 187 },
number = { 47 },
pages = { 1-5 },
doi = { 10.5120/ijca2025925791 },
publisher = { Foundation of Computer Science (FCS), NY, USA }
}
%0 Journal Article
%D 2025
%A Azam Nouri
%T An MLP Baseline for Handwriting Recognition Using Planar Curvature and Gradient Orientation%T
%J International Journal of Computer Applications
%V 187
%N 47
%P 1-5
%R 10.5120/ijca2025925791
%I Foundation of Computer Science (FCS), NY, USA
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 interpretable alternative to convolutional neural networks (CNNs). Using these three handcrafted feature maps as inputs, the curvature–orientation MLP achieves 97%accuracy on MNIST digits and 89%on EMNIST letters. These results underscore the discriminative power of curvature-based representations for handwritten character images and demonstrate that competitive performance is achievable with lightweight, explicitly engineered features.
References
- Cohen, G., Afshar, S., Tapson, J., and van Schaik, A. 2017. EMNIST: An Extension of MNIST to Handwritten Letters. arXiv preprint arXiv:1702.05373.
- Goldman, R. 2005. Curvature Formulas for Implicit Curves and Surfaces. Computer Aided Geometric Design, 22(7), 632–658.
- Kreyszig, E. 1991. Differential Geometry. Dover Publications.
- LeCun, Y., Bottou, L., Bengio, Y., and Haffner, P. 1998. Gradient-Based Learning Applied to Document Recognition. Proceedings of the IEEE, 86(11), 2278–2324.
- Nouri, A. 2025. A Sobel-Gradient MLP Baseline for Handwritten Character Recognition. arXiv:2508.11902.
- Nouri, A. 2025. Curvature-MLP: Code and Training Scripts (v1.0). Zenodo. DOI: 10.5281/zenodo.16934394.
- Siddiqi, I. I., and O’Connor, N. E. 2009. Combining Contour- Based Orientation and Curvature Features for Writer Recognition. In Proc. IWSSIP, 1–4.
- Sobel, I., and Feldman, G. 1968. An Isotropic 3×3 Image Gradient Operator. SAIL Technical Report, Stanford University.
- Srikantan, G., Lam, S. W., and Srihari, S. N. 1996. Gradient- Based Contour Encoding for Character Recognition. Pattern Recognition, 29(7), 1147–1160.
- TensorFlow Datasets Contributors. 2025. Tensor- Flow Datasets (Version 4.9.2). Available at: https: //www.tensorflow.org/datasets.
- Ou, B. 2009. Curvature Function in the Recognition of People’s Handwriting. International Journal of Pure and Applied Mathematics, 52(4), 575–581.
- Zhang, X., and Li, Y. 2023. Learnable Sobel Operators for Edge-Aware Feature Extraction in Convolutional Neural Networks. IEEE Access, 11, 12345–12358.
- Zhang, X., Zhou, Y., and Wu, Y. 2019. Edge Enhancement Network with Learnable Filters for Image Super-Resolution. In Proc. IEEE ICIP, 900–904.
Index Terms
Keywords