Abstract:Stroke extraction and matching are critical for structural interpretation based applications of handwritten Chinese characters, such as Chinese character education and calligraphy analysis. Stroke extraction from offline handwritten Chinese characters is difficult because of the missing of temporal information, the multi-stroke structures and the distortion of handwritten shapes. In this paper, we propose a comprehensive scheme for solving the stroke extraction problem for handwritten Chinese characters. The method consists of three main steps: (1) fully convolutional network (FCN) based skeletonization; (2) query pixel guided stroke extraction; (3) model-based stroke matching. Specifically, based on a recently proposed architecture of FCN, both the stroke skeletons and cross regions are firstly extracted from the character image by the proposed SkeNet and CrossNet, respectively. Stroke extraction is solved by simulating the human perception that once given a certain pixel from non-cross region of a stroke, the whole stroke containing the pixel can be traced. To realize this idea, we formulate stroke extraction as a problem of pairing and connecting skeleton-wise stroke segments which are adjacent to the same cross region, where the pairing consistency between stroke segments is measured using a PathNet [1]. To reduce the ambiguity of stroke extraction, the extracted candidate strokes are matched with a character model consisting of standard strokes by tree search to identify the correct strokes. For verifying the effectiveness of the proposed method, we train and test our models on character images with stroke segmentation annotations generated from the online handwriting datasets CASIA-OLHWDB and ICDAR13-Online, as well as a dataset of Regularly-Written online handwritten characters (RW-OLHWDB). The experimental results demonstrate the effectiveness of the proposed method and provide several benchmarks. Particularly, the precisions of stroke extraction for ICDAR13-Online and RW-OLHWDB are 89.0% and 94.9%, respectively.

Analysis and Extraction of Structural Features of Off-line Handwritten Letters Based on Principal Curves

Chinese Calligraphic Style Representation for Recognition

3D Space Handwriting Recognition Feature Extraction Based on PCA and LDA Combination

Space Handwriting Character Recognition Method Based on Elimination of Connected Stroke

A Structural Method for Online Sketched Symbol Recognition.

Off-Line Handwritten Chinese Character Recognition Based On Double Contour Feature

A Multilevel Approach For Recognizing Characters In Engineering Drawings

Off-line Handwritten Chinese Character Recognition Based on Crossing Line Feature

Face Recognition Based on Curvelets and Local Binary Pattern Features Via Using Local Property Preservation

Palmprint Feature Extraction Based on Curvelet Transform.

Palmprint Feature Extraction Based on Multi-wavelet and Complex Network.

A Novel Method for Vectorizing Historical Documents of Chinese Calligraphy

Chinese character's outline feature and its application

Analysis and Improvement of Directional Element Feature for Off-Line Handwritten Chinese Character Recognition.

On-line Hand-Written Chinese Characters Recognition Method Based on Statistic Structural Features

Research on Automatic Recognition of Unconstratined Handwritten Digits

Segmentation-Driven Offline Handwritten Chinese And Arabic Script Recognition

Baseline-independent feature extraction for Arabic writing

Query Pixel Guided Stroke Extraction with Model-Based Matching for Offline Handwritten Chinese Characters

An Effective Approach to Offline Arabic Handwriting Recognition

Off-line Signature Verification Using Geometric Features Specific to Chinese Handwriting