Chinese Character Recognition Using Non-negative Matrix Factorization

Chinese characters contain three levels of visual information: strokes, radicals, and structures. Stroke, as the basic unit of Chinese characters, is the foundation of the formation of Chinese characters. Some research has shown that strokes can enhance the semantic and orthographic representation in the process of Chinese character acquisition. This shows that the processing of Chinese characters is based on feature processing. This is consistent with the feature processing theory of letter and object recognition. Recent research on visual recognition shows that the recognition of letters and words is based on the termination and the other features (vertices and midsegment) used in object recognition. In this study, we used delayed segment technique on Chinese character recognition to examine the importance of terminations, vertices and midsegment in different masking degree. Experiment 1 selected about 43 high frequency Chinese characters with left-right structures and asked the participants to naming Chinese characters. In experiment 1, we try to examine the importance of terminations, vertices and midsegments based on line unit and consider the degree of masking these features. It was found that the main effect of masking degree was significant. The masking of 35% and 55% of the features makes the RTs significantly longer. The main effect of feature type was significant. Masking vertices and midsegment can make RTs significantly longer. But termination didn’t show its feature effect in experiment 1. When masking these three features (vertices, midsegment and termination) in experiment 1, the features are based on line units, which may destroy the integrity of strokes in Chinese characters. That termination effects couldn’t be obvious in experiment 1 may be caused by this reason. To verify this, in experiment 2, we explore the vertices, midsegment and termination based on stroke units. The results show that the main effect of masking degree was significant, and the RTs were longer by masking 35% and 55% degree. The main effect of feature types was significant. RTs on character recognition are significantly longer by masking the vertices, midsegment and termination. This shows that the vertices, midsegment and termination based on stroke unit are also important in Chinese character recognition. Based on the above research results, it can be seen that the key features on Chinese character recognition are in line with the neuronal recycling hypothesis. According to neuronal recycling hypothesis, human’s ability to learn from culture depends on the retrieval of preexisting brain circuits, and the key features in word recognition may be derived from our key features in object processing. The hook is considered to be a special termination in Chinese characters. The results of experiment 2 also emphasized the key role of hook features in character recognition.

Chinese characters are constructed by basic strokes based on structural rules. In handwritten characters, the shapes of the strokes may vary to some extent, but the spatial relations and geometric configurations of the strokes are usually maintained. Therefore these spatial relations and configurations could be regarded as invariant features and could be used in the recognition of handwritten Chinese characters. In this paper, we investigate the structural knowledge in Chinese characters and propose the stroke spatial relationship representation (SSRR) to describe Chinese characters. An On-Line Chinese Character Recognition (OLCCR) method using the SSRR is also presented. With SSRR, each character is processed and is represented by an attribute graph. The process of character recognition is thereby transformed into a graph matching problem. After careful analysis, the basic spatial relationship between strokes can be characterized into five classes. A bitwise representation is adopted in the design of the data structure to reduce storage requirements and to speed up character matching. The strategy of hierarchical search in the preclassification improves the recognition speed. Basically, the attribute graph model is a generalized character representation that provides a useful and convenient representation for newly added characters in an OLCCR system with automatic learning capability. The significance of the structural approach of character recognition using spatial relationships is analyzed and is proved by experiments. Realistic testing is provided to show the effectiveness of the proposed method.

Chinese Optical Character Recognition (OCR) technology is essential for digital transformation in Chinese regions, enabling automated document processing across various applications. However, Chinese OCR systems struggle with visually similar characters, where subtle stroke differences lead to systematic recognition errors that limit practical deployment accuracy. This study develops FLIP (Feedback Learning-based Intelligent Plugin), a lightweight post-processing plugin designed to improve Chinese OCR accuracy across different systems without external dependencies. The plugin operates through three core components as follows: UTF-8 encoding-based output parsing that converts OCR results into mathematical representations, error correction using information entropy and weighted similarity measures to identify and fix character-level errors, and adaptive feedback learning that optimizes parameters through user interactions. The approach functions entirely through mathematical calculations at the character encoding level, ensuring universal compatibility with existing OCR systems while effectively handling complex Chinese character similarities. The plugin’s modular design enables seamless integration without requiring modifications to existing OCR algorithms, while its feedback mechanism adapts to domain-specific terminology and user preferences. Experimental evaluation on 10,000 Chinese document images using four state-of-the-art OCR models demonstrates consistent improvements across all tested systems, with precision gains ranging from 1.17% to 10.37% and overall Chinese character recognition accuracy exceeding 98%. The best performing model achieved 99.42% precision, with ablation studies confirming that feedback learning contributes additional improvements from 0.45% to 4.66% across different OCR architectures.

This study designed a finger-worn device-named the Chinese FingerReader-that can be practically applied by visually impaired users for recognizing traditional Chinese characters on the micro internet of things (IoT) processor. The device is portable, easy to operate, and designed to be worn on the index finger. The Chinese FingerReader on the index finger contains a small camera and buttons. The small camera captures images by identifying the relative position of the index finger to the printed text, and the buttons are applied to capture an image by visually impaired users and provide the audio output of the corresponding Chinese character by a voice prompt. To recognize Chinese characters, English letters, and numbers, a robust Chinese optical character recognition (OCR) system was developed according to the training strategy of an augmented convolution neural network algorithm. The proposed Chinese OCR system can segment a single character from the captured image, and the system can accurately recognize rotated Chinese characters. The experimental results revealed that compared with the OCR application programming interfaces of Google and Microsoft, the proposed OCR system obtains 95% accuracy rate in dealing with rotated character images where the Google and Microsoft OCR APIs only obtain 65% and 34% accuracy rates. These results illustrate that the proposed OCR system was more suitable for the needs of visually impaired people in actual use. Finally, three usage scenarios were simulated, and the accuracy and operational performance of the system were tested. Field tests of this system were conducted for visually impaired users to verify its feasibility.

Scene character recognition problem has attracted a great attention in computer vision field. As one of the most widely used characters, Chinese characters are more complicated than European characters, especially when it comes to those characters appeared in various scene texts. The recognition of Chinese scene characters is still far from a satisfactory level owing to the lack of adequate training data and efficient learning methods. In this paper, we propose a new strategy to divide training data and combine multiple classifiers to get a better performance of Chinese scene character recognition. Besides, facing the problem of data shortage, we propose a scene character synthesis method to gain enough training data. By applying the proposed training strategy, the average recognition accuracies of Random Forest (RF) and Support Vector Machine (SVM) have been improved by nearly 22% and 14%, respectively.

In this study, we investigated the role of visual and phonological information in lexical access of Chinese characters. Homophonic English words have been the main source of stimuli for word recognition research. However, since these stimuli also often look alike, visual and phonological information may be confounded in reported experiments. In contrast, many homophonic Chinese characters are visually distinct. In addition, visually similar characters often have very different pronunciations. These characteristics allow a more controlled investigation of the roles of visual and phonological information in activation of meaning. In the present study, two types of Chinese characters were used in a semantic categorization paradigm: integrated characters, which contain strokes that are not separable; and compound characters, which contain at least two clearly identifiable components. The results show that the recognition of a Chinese integrated character depends primarily on visual information, whereas the recognition of a Chinese compound character relies on visual, phonological, and semantic information. It is concluded that visual information plays a greater role in Chinese character recognition than has previously been documented.

This paper applies deep siamese network to one-shot Chinese handwritten character recognition. Different from common image classification tasks, the CASIA HWDB1.1 dataset used here contains more than 3000 categories, with only few samples in each one. We propose a basic deep siamese model as well as an improved model with multi-layer features mechanism and batch normalization for extracting the similarity of the input pairs, and implement one-shot recognition by categorizing the test example to the class where the support sample is the most similar. Experiments prove that our model is able to recognize Chinese characters of unseen classes in training with only one support example efficiently.

Handwritten Chinese character recognition (HCCR) is an important research field of pattern recognition, which has attracted extensive studies during the past decades. Recently convolutional neural network (CNN) based methods have achieved the state-of-the-art performance for handwritten Chinese character recognition. Nevertheless, handwritten Chinese character recognition is still limited to be effectively used in the actual environment due to the large-scale vocabulary and great diversity of handwriting style. In this paper, we constructed a handwritten Chinese character recognition service based on convolutional neural network, which tries to make effective use of handwritten based printed fonts and existing handwritten database. At the same time, the service can effectively collect more handwritten data to expand the training dataset, which makes it easy to adapt to the new handwriting styles. Meanwhile, We propose a multi-level recognition theory applied to online handwritten Chinese character recognition, which may improve the accuracy of handwritten Chinese character recognition and break the limitations of handwritten Chinese character recognition by identifying the structure of Chinese characters and possible stroke orders firstly. Furthermore, we try to apply the method of online character recognition to the offline character recognition based on the basic writing rules.

In this paper we study the problem of Chinese character recognition in video. We propose a series of algorithms on Chinese character division, tracking. Based on them we design a multi-level sorter. Firstly we extract the features of some samples and employ K-means clustering algorithm to carry on I level classification. Secondly, we employ the algorithm of multi back propagation neural network (MBPNN) to classify every category once again and we call it II level classification. Finally, we carry on the experiment and the testing result proves that these algorithms are effectively and recognition rate is higher than conventional back propagation neural network.

Stroke extraction of Chinese characters plays an important role in Chinese character recognition, calligraphy aesthetic quality evaluation and handwriting synthesis. However, most of the existing methods perform well on printed characters, but not well on ancient calligraphy and handwritten characters. In addition, most of these methods only use a single model based on graph theory or deep learning to achieve end-to-end stroke extraction, resulting in limited room for model performance improvement. In this paper, we propose a Chinese character stroke extraction method based on hierarchical multi-model including four steps, which apply KNN clustering algorithm into stroke extraction and improved the performance of stroke extraction based on instance segmentation.Firstly, we use a stacked model driven calligraphic character recognition method to recognize the input character image. Then we use the font recognition model to match the input character image with images with annotated strokes. For those images with matching rate higher than 99\%, stroke extraction method based on KNN pixel clustering is used to extract the strokes. Thirdly, for those images with matching rate less than 99\%, we separate the non-connected regions of the input character image according to the graphic attributes to find out the independent strokes. Finally, the instance segmentation-based stroke extraction method is applied to extract each individual stroke. Experimental results show that the average accuracy of the proposed method on the five test subsets of CSSCD, LTH, SS, HLJ, FZJTJW and FZLBJW is 94.3\%, 96.2\%, 91.7\%, 92.9\% and 89.4\%, respectively, which is greatly improved compared with the existing methods, which proves the effectiveness of the method described in this paper. Our code and datasets are publicly available at \url{https://github.com/JacksonMa618/MLStroke}

The present study adopts the priming word-judging paradigm to investigate whether the critical period will influence radical and lexical tone awareness during the bilinguals' Chinese character recognition. Two types of bilinguals (simultaneous and sequential bilinguals) will be recruited as participants in the present study. The different features between these two types of participants are used to test the influence of the critical period. Through a Chinese character decision task experiment, we hypothesis that the critical period will influence the semantic radical and lexical tone awareness while processing the Chinese characters of bilinguals. In addition, this study will provide some advice to learning Chinese characters in second language acquisition.

Due to the various individual handwriting styles and diverse structures of Chinese characters, it is rather intricate to recognize handwritten Chinese characters precisely. In this paper, a muti-stage system that combines a Differentiable Binarization Network (DBNet) for text detection and Scene Text Recognition with a Single Visual Model (SVTR) for effective character recognition is developed, aiming to offer a potential solution for correct identification of handwritten Chinese characters in real-world scenarios. More specifically, the proposed system utilized the CASIA-HWDB2.x dataset from the Institute of Automation of the Chinese Academy of Sciences to train text detection and recognition models. For the detection, the Differentiable Binarization method in the Head Network applies a dynamic threshold map for text-background differentiation. On top of the SVTR employed as a recognition module, this research proposed a performance-optimized PP-LCNet architecture and the Guided Training of Continuous Time-Series Classifiers (GTC) with TextConAug, which is a data augmentation strategy to accommodate real-world scenarios. Briefly, DBNet and SVTR are integrated to form an end-to-end system. The output of DBNet after undergoing cropping and direction classifier, guides SVTR for a comprehensive system, ensuring seamless detection and recognition of handwritten Chinese characters. Experimental results indicate that this system achieves a promising performance and can provide accurate information on handwritten Chinese characters.

This paper proposes a hidden Markov mesh random field (HMMRF)-based stochastic model for off-line handwritten Chinese characters recognition using statistical observation sequences embedded in the strokes of a character. Due to the great amount of Chinese characters and many different writing styles or variations, the recognition of handwritten Chinese characters becomes more difficult and challenging than any other character recognition. In our approach, a new framework based on HMMRF model is put forward at first. The estimation of model parameters and state sequence decoding algorithms are also discussed later. Besides the mathematical model and corresponding issues, nonlinear shape normalization scheme that modifies the distortion and adjusts the correlation of strokes is applied. Two types of stroke-based features are extracted for the rough classification and observation sequence respectively. Experimental results on isolated handwritten Chinese characters demonstrate the effectiveness of our approach.

Chinese characters presented tachistoscopically to young readers of minimal reading ability and to young readers of better ability produce reliable differences in recognition thresholds in favor of the good readers. Simultaneous presentations of compared characters also reduces perceptual thresholds as compared to successive presentation with minimal readers being inferior to good readers on the successive task. Chinese characters have been used as intermediaries in reading and improved performance attributed to the fact that the Chinese characters are related to words as units rather than to phonemic elements as in the Roman alphabet. However, since visual factors also appear to be involved in recognition of Chinese characters in a manner distinguishing between minimal and good young readers, absence of phonemic word structure may not be responsible for the improvement arising from substitution of the Chinese for Roman script.

As the development of three-dimensional interaction, in-air writing arises as a novel interaction method, and brings traditional writing behavior to the 3D space. In this paper, we propose a discriminative three-level classifier for in-air handwritten Chinese character recognition. Since the existence of characters with similar shapes and structures results in one of the main difficulties of Chinese handwritten character recognition, this paper tackles the problem caused by similar characters using two discriminative analysis techniques. First, we exploit the learning vector quantization (LVQ) to obtain discriminative prototypes. Then, we use the adaptive discriminative locality alignment (ADLA) to distinguish candidate character class among similar character classes. In the experiments, we evaluate the proposed method on the IAHCC-UCAS2014 dataset constructed by ourselves using the writing-in-the-air system based on the Leap Motion Controller. The experimental results show that the proposed method obtain higher recognition accuracy with lower computational cost.