Rotation-free Online Handwritten Character Recognition Using Linear Recurrent Units

TL;DR

This paper introduces a rotation-invariant online handwritten character recognition method using Sliding Window Path Signature features and Linear Recurrent Units, achieving high accuracy even with large rotational deformations.

Contribution

The work presents a novel combination of SW-PS features and lightweight LRU classifiers for rotation-free online handwritten character recognition, improving robustness and efficiency.

Findings

Achieved over 99.6% accuracy on digit recognition with rotation up to ±180°

Surpassed competing models in convergence speed and accuracy

Effective on multiple character datasets including digits, letters, and radicals

Abstract

Online handwritten character recognition leverages stroke order and dynamic features, which generally provide higher accuracy and robustness compared with offline recognition. However, in practical applications, rotational deformations can disrupt the spatial layout of strokes, substantially reducing recognition accuracy. Extracting rotation-invariant features therefore remains a challenging open problem. In this work, we employ the Sliding Window Path Signature (SW-PS) to capture local structural features of characters, and introduce the lightweight Linear Recurrent Units (LRU) as the classifier. The LRU combine the fast incremental processing capability of recurrent neural networks (RNN) with the efficient parallel training of state space models (SSM), while reliably modelling dynamic stroke characteristics. We conducted recognition experiments with random rotation angle up to $\pm 180^{\circ}$ on three subsets of the CASIA-OLHWDB1.1 dataset: digits, English upper letters, and Chinese radicals. The accuracies achieved after ensemble learning were $99.62\%$, $96.67\%$, and $94.33\%$, respectively. Experimental results demonstrate that the proposed SW-PS+LRU framework consistently surpasses competing models in both convergence speed and test accuracy.