Abstract:The flourishing blossom of deep learning has witnessed the rapid development of Chinese character recognition. However, it remains a great challenge that the characters for testing may have different distributions from those of the training dataset. Existing methods based on a single-level representation (character-level, radical-level, or stroke-level) may be either too sensitive to distribution changes (e.g., induced by blurring, occlusion, and zero-shot problems) or too tolerant to one-to-many ambiguities. In this paper, we represent each Chinese character as a stroke tree, which is organized according to its radical structures, to fully exploit the merits of both radical and stroke levels in a decent way. We propose a two-stage decomposition framework, where a Feature-to-Radical Decoder decomposes each character into a radical sequence and a Radical-to-Stroke Decoder further decomposes each radical into the corresponding stroke sequence. The generated radical and stroke sequences are encoded as a radical-structured stroke tree (RSST), which is fed into a Tree-to-Character Translator based on the proposed Weighted Edit Distance to match the closest candidate character in the RSST lexicon. We have conducted extensive experiments on various datasets, such as handwritten, printed artistic, scene character datasets. The experimental results demonstrate that the proposed method outperforms the state-of-the-art single-level methods by increasing margins as the distribution difference becomes more severe in the blurring, occlusion, and zero-shot scenarios. For example, compared with the previous SOTA method, our method improve performance by 1.74–7.58% in the handwritten character zero-shot settings.

Context driven chinese string segmentation and recognition

A segmentation algorithm for handwritten Chinese character strings

Segmentation of On-Line Handwritten Chinese Characters Based on Stroke Merging and Dynamic Programming

A hidden Markov model based segmentation and recognition algorithm for Chinese handwritten address character strings

A Novel Segmentation and Recognition Algorithm for Chinese Handwritten Address Character Strings

Multi-Step Segmentation Method Based on Adaptive Thresholds for Chinese Calligraphy Characters.

A Novel Short Merged Off-line Handwritten Chinese Character String Segmentation Algorithm Using Hidden Markov Model

An approach for handwritten Chinese text recognition unifying character segmentation and recognition

Substring Alignment Method for Lexicon Based Handwritten Chinese String Recognition and Its Application to Address Line Recognition

Handwritten Chinese Address Segmentation and Recognition Based on Merging Strokes

Recognition of Handwritten Chinese Text by Segmentation: A Segment-annotation-free Approach

Post Processing for Offline Chinese Handwritten Character String Recognition

On-line Handwritten Chinese Word Recognition Based on Lexicon

Chinese character recognition with radical-structured stroke trees

On-line Chinese Character Recognition System for Overlapping Samples

Chinese Character Components Segmentation Method Based on Faster RCNN

Handwritten Chinese Address Segmentation and Recognition Based on Structural Matching

Application of bi-gram driven chinese handwritten character segmentation for an address reading system

LCSegNet: An Efficient Semantic Segmentation Network for Large-Scale Complex Chinese Character Recognition

Graph Model Optimization Based Historical Chinese Character Segmentation Method

Handwritten Chinese/Japanese text recognition using semi-Markov conditional random fields