Research on Cross Feature Code Extraction of Handwritten Character

Youwei Du

2003-01-01

Abstract:By research on traditional way of cross feature mode extraction, this paper introduced an improved algorithm of changing polar coordinates origin.This algorithm was good for 6,9,B,E,P.It was based on computer programming.Expected result was achieved by using this algorithm.

What problem does this paper attempt to address?

Chinese Handwriting Identification Based on Stable Spectral Feature of Texture Images

Yuchen Yan,Qinghu Chen,Wei Deng,Feng Yuan

DOI: https://doi.org/10.22266/ijies2009.0331.03

2009-01-01

International Journal of Intelligent Engineering and Systems

Abstract:Abstrac t: In this paper, a spectral feature extraction method based on fast Fourier transformation for writer identifi cation is presented. According to the constructibility of the texture image of handwriting, we put forward an estimation method for mathematical expectation value of texture image’s spectral features. This method eliminates the randomness of spectral features and gets stable spectral features. The experiment results show that this approach enhances the identification accuracy to a large extent using data sets with large handwriting samples.
Space Handwriting Character Recognition Method Based on Elimination of Connected Stroke

LAI Ying-chao,ZENG Jian-ming,SHEN Hai-bin

DOI: https://doi.org/10.3969/j.issn.1000-3428.2012.19.040

2012-01-01

Abstract:During the process of continuous collection of information in 3D space handwriting,connected strokes of part of characters appeares,which affectes the rate of characters recognition.To solve this problem,an approach of 3D space handwriting character recognition based on elimination of connected stroke is represented.3D space characters are flattened.Feature extraction of turning points and direction of strokes is performed.To avoid eliminating the wrong strokes,Support Vector Machine(SVM) is adopted to divide the unknown characters into 2 sections: connected-stroke characters and non-connected-stroke characters.The connected strokes are eliminated by using stroke writing features which put the trajectory of a 3D space character into a flat trajectory of the character.Character recognition is performed by using flat character classifier directly.Experimental results show that the effect of eliminating connected stroke using this approach is remarkable and a higher recognition rate can be obtained by using existed character library.
AN ALGORITHM FOR EXTRACTING CHARACTER FEATURE POINTS BASED ON CIRCULAR TEMPLATE

Cao Zhonghua,Guo Bin,Wang Yan

DOI: https://doi.org/10.3969/j.issn.1000-386X.2011.03.078

2011-01-01

Abstract:This paper presents a new algorithm for extracting character feature points based on circular template.By analysing the background region numbers and relevant characteristics of the circular template divided by the character strokes,the algorithm determines the feature point position of the character.Then the fork position is exactly located using maximum circle criterion.Experiments show that the algorithm is fast,accurate and robust in extracting the feature point information.
Improvements on EMG-based handwriting recognition with DTW algorithm.

Chengzhang Li,Zheren Ma,Lin Yao,Dingguo Zhang

DOI: https://doi.org/10.1109/EMBC.2013.6609958

2013-01-01

Abstract:Previous works have shown that Dynamic Time Warping (DTW) algorithm is a proper method of feature extraction for electromyography (EMG)-based handwriting recognition. In this paper, several modifications are proposed to improve the classification process and enhance recognition accuracy. A two-phase template making approach has been introduced to generate templates with more salient features, and modified Mahalanobis Distance (mMD) approach is used to replace Euclidean Distance (ED) in order to minimize the interclass variance. To validate the effectiveness of such modifications, experiments were conducted, in which four subjects wrote lowercase letters at a normal speed and four-channel EMG signals from forearms were recorded. Results of offline analysis show that the improvements increased the average recognition accuracy by 9.20%.
Handwriting Chinese Character Computer Writer Identification Based on Stroke——Computer Writer Identification Based on Motion Feature,First and Last Feature

Sai Wu

2008-01-01

Abstract:In order to overcome the defect of the previous research based on the text-dependent,the research defined the motion fea-ture first and last feature as feature space based on the basic stroke of handwriting Chinese characters. The experiment based on the 4 basic strokes which extracted from the 70 Chinese characters of 10 writers. The research makes a satisfied result.
A Novel Method for Vectorizing Historical Documents of Chinese Calligraphy

Junsong Zhang,Hongwei Lin,Jinhui Yu

DOI: https://doi.org/10.1109/cadcg.2007.4407884

2007-01-01

Abstract:We develop a novel method for feature point detection and employ it to generate outline font from historical document of Chinese calligraphy. The feature points at a character contour subdivide the contour into segments. Each segment can be then fitted by a parametric curve to obtain the outline font. Some experimental results are also presented in the paper.
Algorithm to Segment and Recognize Handwritten Chinese Characters Using Address Interpretation

王嵘,丁晓青,刘长松

DOI: https://doi.org/10.3969/j.issn.1000-3428.2004.07.015

2004-01-01

Abstract:In order to segment the address characters overlapped or crossed in strokes, a stroke extract and merge based method is used. And after the dynamic programming processing with the address interpretation, the final segmentation result and the sorting area are interpreted. An experiment on 443 real handwritten address lines is performed. The results show that a correct rate of the sorting area including province and city is up to 66%.
Exploiting Gabor Feature Extraction Method for Chinese Character Writing Quality Evaluation

Zhixiao Wang,Wenyao Yan,Mingtao Guo,Jiulong Zhang

DOI: https://doi.org/10.1088/1742-6596/1575/1/012065

2020-06-01

Journal of Physics: Conference Series

Abstract:Abstract The automatic evaluation of Chinese character writing quality has a wide application prospect. Most of the existing evaluation methods of Chinese character writing quality are based on radical segmentation and feature judgment, which require the high accuracy of Chinese character segmentation. However, there are many problems in the real handwriting, such as continuous writing, uneven strength of writing, personalized writing style and so on, which lead to the difficulty of segmentation in the ordinary handwriting. To solve the above problems, we propose an effective method based on image texture where the uniformity of writing lines and writing style is taken as an effective criterion. In our method, Gabor transform is used to extract the image features of writing samples, and finally the statistical learning method of support vector machine is used to effectively evaluate the writing quality. Experiments on multiple real datasets including CHAED show that our method is effective and accurate. The advantage of this method is that it does not need to segment fonts, and the cost of global feature extraction is small.
Robust Feature Extraction For Character Recognition Based On Binary Images

Lijun Wang,Li Zhang,Yuxiang Xing,Zhiming Wang,Hewei Gao

DOI: https://doi.org/10.1117/12.650386

2006-01-01

Abstract:Optical Character Recognition (OCR) is a classical research field and has become one of most successful applications in the area of pattern recognition. Feature extraction is a key step in the process of OCR. This paper presents three algorithms for feature extraction based on binary images: the Lattice with Distance Transform (DTL), Stroke Density (SD) and Co-occurrence Matrix (CM). DTL algorithm improves the robustness of the lattice feature by using distance transform to increase the distance of the foreground and background and thus reduce the influence from the boundary of strokes. SD and CM algorithms extract robust stroke features base on the fact that human recognize characters according to strokes, including length and orientation. SD reflects the quantized stroke information including the length and the orientation. CM reflects the length and orientation of a contour. SD and CM together sufficiently describe strokes. Since these three groups of feature vectors complement each other in expressing characters, we integrate them and adopt a hierarchical algorithm to achieve optimal performance. Our methods are tested on the USPS (United States Postal Service) database and the Vehicle License Plate Number Pictures Database (VLNPD). Experimental results shows that the methods gain high recognition rate and cost reasonable average running time. Also, based on similar condition, we compared our results to the box method proposed by Hannmandlu [18]. Our methods demonstrated better performance in efficiency.
Off-line Handwritten Chinese Character Recognition Based on Crossing Line Feature

Youbin Chen,Xiaoqing Ding,Youshou Wu

DOI: https://doi.org/10.1109/ICDAR.1997.619842

1997-01-01

Abstract:A new method to extract crossing line feature for off-line handwritten Chinese character recognition is proposed in this paper. Firstly, the input pattern is nonlinearly normalized in order to compensate for shape variations. Secondly, the normalized pattern is separated into four subpatterns according to the four kinds of elementary strokes. Thirdly, the four subpatterns are uniformly divided into MX M cells respectively. In every cell, the crossing lines are counted. Then a 4M 2 dimensional feature vector is generated. An off-line handwritten Chinese character recognition system is built based on this feature. Our experiments have demonstrated the effectiveness of the method proposed in this paper.
Analysis and Improvement of Directional Element Feature for Off-Line Handwritten Chinese Character Recognition.

YB Chen,YS Wu,JR Beveridge

DOI: https://doi.org/10.1117/12.304620

1998-01-01

Abstract:Past work has shown the directional element feature (DEF) to be useful for off-line recognition of handwritten Chinese characters. This paper presents refinements to this approach that significantly improve recognition performance. These refinements do the following: 1) assign fuzzy attributes to stroke edge pixels, 2) divide the character image into fuzzy membership cells, 3) smooth saw-toothed edges, 4)enhance information at the boundary of the character image, and 5) enhance horizontal and vertical stroke edges. The first two compensate for variations in stroke position, length, inclination and width. Smoothing can correct some aliasing and dropout problems in the character images. The latter two emphasize the more important aspects of a character image. All refinements improve recognition, and when used together, they increase performance by about 10%: from 84.15% to 94.09% on a set of 3,755 Chinese characters. Experiments over all subsets of refinements are included, While not all refinements lead to equal improvement, all are necessary to achieve the highest level of recognition.
Research Of 3d Chinese Calligraphic Handwriting Recur System And Its Key Algorithm

Ying-Fei Wu,Yue-Ting Zhuang,Yun-He Pan,Jiang-Qin Wu,Fei Wu

DOI: https://doi.org/10.1007/978-3-540-71457-6_10

2007-01-01

Abstract:Chinese calligraphy is a precious Chinese art. It is pictographic and each calligraphist has his own writing style. Even people whose native language is Chinese will have difficulties in writing a demanded beautiful calligraphy style, not to say people who know little about Chinese calligraphy. In order to help people enjoy the art of calligraphy writing and find out the process how it was written, we implement a new approach to animate its writing process by 3D visualization method. In the paper two novel algorithms are also presented to extract the intrinsic feature which is needed to rebuild the writing process: 1) extract strokes order from an offline Chinese calligraphic handwriting; 2) estimate varied stroke's thickness. Finally, experiment result is given to demonstrate the application.
Research on Translation and Rotation in Handwritten Digits Recognition

Huan Wang,Yi Wan

DOI: https://doi.org/10.1109/icomssc45026.2018.8941675

2018-09-01

Abstract:Handwritten digits recognition is a classical problem in machine learning and neural networks. It plays an important role in pattern recognition. In the traditional BP Neural Network, the recognition accuracy rate will drop significantly while translating or rotating test images. However, when extracting the features of the original images as the input in which the neural network is most interested, the performance of the network is improved. This paper proposes two new methods of processing the original images and extracting the feature. One is to add a sliding window and translate appropriate units to center images. The other is to rotate the input image to align its axes to its principle directions. This is achieved by Principal Component Analysis(PCA) to the input image and find its principle directions. According to the experiment results, the problem as above has been solved, and the recognition accuracy rate has improved from 96.66% to 97.12%. It is proved that the methods are simple but effective.
An Improved Method for Similar Handwritten Chinese Character Recognition

Fang Yang,Xue-Dong Tian,Xin Zhang,Xin-Biao Jia

DOI: https://doi.org/10.1109/iitsi.2010.70

2010-01-01

Abstract:Similar characters existed in Chinese character affect much on enhancing the handwritten Chinese character recognition rate. An improved method is presented in this paper, which combines structural and statistical features for similar handwritten Chinese character recognition,. Four-corner code feature that based on stroke structure is used to get the similar character set. According to the different strokes on the four corners, four-corner code feature can dispatch some characters which are similar in shape into different similar sets. Statistical hierarchy contour features are extracted from the characters in the same similar set, then support vector machine (SVM) is adopted as classifier to recognize the similar characters. This method reduces the complexity of recognizing similar characters, and the experiment results on common used 500 Chinese characters show the effectiveness.
On-Line Sketch Recognition Using Direction Feature

Wei Deng,Lingda Wu,Ronghuan Yu,Jiazhe Lai

DOI: https://doi.org/10.1007/978-3-642-40477-1_16

2013-01-01

Abstract:Sketch recognition is widely used in pen-based interaction, especially as the increasing popularity of devices with touch screens. It can enhance human-computer interaction by allowing a natural/free form of interaction. The main challenging problem is the variability in hand drawings. This paper presents an on-line sketch recognition method based on the direction feature. We also present two feature representations to train a classifier. We support our case by experimental results obtained from the NicIcon database. A recognition rate of 97.95% is achieved, and average runtime is 97.6ms using a Support Vector Machine classifier.
Query Pixel Guided Stroke Extraction with Model-Based Matching for Offline Handwritten Chinese Characters

Tie-Qiang Wang,Xiaoyi Jiang,Cheng-Lin Liu

DOI: https://doi.org/10.1016/j.patcog.2021.108416

IF: 8

2022-03-01

Pattern Recognition

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.

computer science, artificial intelligence,engineering, electrical & electronic
Calligraphy Stroke Extraction in CADAL.

Kai Yu,Zhen-ming Yuan,Jiangqin Wu

DOI: https://doi.org/10.1007/978-3-642-34752-8_57

2012-01-01

Abstract:Stroke is the basic unit of Chinese characters. Stroke extraction is the key procedure for calligraphy reconstruction in modern digital libraries, which faces the difficulties caused by deformation and tedious manual work. A novel stroke extraction approach for Chinese calligraphy character is proposed in this paper. Firstly, the corner points on the contours are detected with fast-CPDA, and a half-edge coloring graph is presented to help connecting the contour segments on the same side of a stroke. Then the connected contour segments on opposite sides are matched to form the whole strokes. Finally, the results are refined by rule-based verification. Experiments show that the proposed method can extract basic strokes for simple calligraphy characters automatically and help stroke extraction for complicated calligraphy characters with little manual work. © 2012 Springer-Verlag.
Research of Chinese Handwritten Text Segmentation Algorithm

Yingyong Zou,Yongde Zhang,Xinyan Cao,Guangbin Yu

DOI: https://doi.org/10.14257/ijmue.2014.9.12.12

2014-01-01

International Journal of Multimedia and Ubiquitous Engineering

Abstract:OCR is a complicated process, there are many factors that can influence the recognition rate.Early period people tried to optimize the classifier to obtain high recognition rate, but the premise is that there is only one character no matter print or handwritten.For the performance of classifier has been promoted a lot, recognition rate for single character is high enough for commercial use.With the development of the demand for handwritten text recognition, how to raise the recognition rate of OCR system becomes very important.Unlike OCR system for print which focus on classifier.The research of OCR system for handwritten text is mainly on character segmentation.Statistical analysis showed that the mistake made by missegment is more than the mistake made by classifier.This is decided by the feature of handwritten text.There are more randomness and the lines are not horizontal, besides that, handwritten Chinese characters are more like overlapped and the gaps between characters are smaller.So this is the difficulty of handwritten Chinese characters.In this paper, the mutil-step searching nonlinear line exaction algorithm the paper proposed is easy and the accuracy is high, which can tackle the some weaknesses of direct projection method and indirect projection.
Image iterative method for handwritten Chinese character recognition

Wanbo Yu,Yaosheng Li,Hongwei Peng,Li Zhang

DOI: https://doi.org/10.1088/1742-6596/1684/1/012101

2020-11-01

Journal of Physics: Conference Series

Abstract:Abstract Although accuracy and calculation time of handwritten Chinese character recognition have been greatly improved, it has not yet reached the practical application standard. Recently literatures, iterative method is used to recognize face images and good results are obtained. Therefore, image recognition is improved based on this idea. In this paper, linear combination of trigonometric functions is used as auxiliary function and image to construct discrete dynamic system. By adding bevel and shift to construct font surface and so on, the system structure is improved to solve problem of system convergence caused by large area flatness of handwritten Chinese character image, and good recognition effect is achieved. Preliminary experiments(for example, extracting 20 Chinese characters written by 9 people in the data set) are carried out, and recognition rate can reach 100% when all of them are trained, and recognition rate can reach 76.7% when each Chinese character is trained to 3 pieces; in addition, experiments were conducted on handwritten Chinese character sets changing from different angles, and the recognition rate reached 80%, which was compared with other algorithms to verify the feasibility of handwritten Chinese character recognition algorithm based on chaotic iterative trajectory characteristics of images.
Writer identification using edge based features

Fan Zhenyin,Guo Zhenhua,Chen Youbin

DOI: https://doi.org/10.1109/ACPR.2015.7486537

2015-01-01

Abstract:In this paper we present a new method for writer identification, which extract original Local Binary Pattern(LBP) of different radius and Edge descriptors from the edge points of the handwriting. Then, we make combinations of these edge based features. Experimental results demonstrate that the combination of edge points based features outperform traditional features extracted from the whole text, which can get state-of-the-art performance on CVL andICDAR2013 datasets.

Research on Cross Feature Code Extraction of Handwritten Character

Chinese Handwriting Identification Based on Stable Spectral Feature of Texture Images

Space Handwriting Character Recognition Method Based on Elimination of Connected Stroke

AN ALGORITHM FOR EXTRACTING CHARACTER FEATURE POINTS BASED ON CIRCULAR TEMPLATE

Improvements on EMG-based handwriting recognition with DTW algorithm.

Handwriting Chinese Character Computer Writer Identification Based on Stroke——Computer Writer Identification Based on Motion Feature,First and Last Feature

A Novel Method for Vectorizing Historical Documents of Chinese Calligraphy

Algorithm to Segment and Recognize Handwritten Chinese Characters Using Address Interpretation

Exploiting Gabor Feature Extraction Method for Chinese Character Writing Quality Evaluation

Robust Feature Extraction For Character Recognition Based On Binary Images

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

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

Research Of 3d Chinese Calligraphic Handwriting Recur System And Its Key Algorithm

Research on Translation and Rotation in Handwritten Digits Recognition

An Improved Method for Similar Handwritten Chinese Character Recognition

On-Line Sketch Recognition Using Direction Feature

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

Calligraphy Stroke Extraction in CADAL.

Research of Chinese Handwritten Text Segmentation Algorithm

Image iterative method for handwritten Chinese character recognition

Writer identification using edge based features