Handwritten Chinese Character Blind Inpainting with Conditional Generative Adversarial Nets

Forecasting short-term, nonrecurring traffic dynamics caused by incidents is an essential capability in the Intelligent Transportation Systems. This research proposes a prediction framework in which Conditional Deep Convolutional Generative Adversarial Nets (C-DCGAN) is trained to predict the traffic spillbacks patterns associated with freeway incidents at merging bottleneck. Speed tensors, which depict the spatiotemporal incident-induced impacts for multiple neighboring routes, is a suitable object for the GAN model to understand and predict. Further, we demonstrated how to use the mesoscopic Dynamic Traffic Assignment (DTA) model DynusT to generate a large number of training data, thus speeding up the model training. The developed model achieves both statistical and spatial similarities between predicted speed tensors and actual tensors, to 83.84%. To the best of our knowledge, this line of work is one of the first attempts in the literature to train the Machine Learning model to predict speed tensor representation of multi-location incident-induced spatiotemporal impact at merging bottleneck and speeding up the training via simulation.

This paper describes a new feature extraction method contributing to improvement of the performance of a handwritten Chinese character recognition system. By using enhanced weighted dynamic meshes based on nonlinear normalization, this method not only avoids the zigzags and other undesirable side effects introduced in the original Yamada et al.’s nonlinear normalization method but also avoids additional feature normalization process in the original Lian-Wen Jin et al.’s and WU Tian-lei et al.’s dynamic mesh method. Experiment on HCL2000, a handwritten Chinese character database, shows that our method achieves superior performance.KeywordsChinese CharacterCharacter RecognitionFeature Extraction MethodDynamic MeshHandwritten CharacterThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Today, more and more foreigners go to China and are emerged into studying Chinese. Thereinto, how to write Chinese characters is a very important and difficult task. As computers and internets develop, many teachers for Chinese Education want to use pattern recognition technologies to automatically evaluate and direct the quality of Chinese characters written by foreign students through document scanning. Actually, this is a handwritten character evaluation and identification problem. In this paper, we investigate and compare several character identification methods for Chinese Education within a classification framework. First, some two-class classification techniques with different features and classifiers (BP neural networks and SVMs) are investigated to identify each handwritten Chinese character. Moreover, in character identification, positive examples are always conjunctive, but negative examples are diffused in most cases. Consequently, we use one-class classification technique (one-class SVMs) to perform this handwritten character identification. In order to overcome the sensitivity to the SVM parameters, we propose a variant one-class SVM system - Bagged One-Class SVMs, which integrate many one-class SVMs with sample bagging. Some experiments of evaluating real handwritten Chinese characters by foreigners are performed, which show that general handwritten character identification is a big challenge and one-class classification technique is a potential researching and developing direction.

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.

The task of handwritten Chinese character recognition is one of the most challenging areas of human handwriting classification. The main reason for this is related to the writing system itself which encompasses thousands of characters, coupled with high levels of diversity in personal writing styles and attributes. Much of the existing work for both online and off-line handwritten Chinese character recognition has focused on methods which employ feature extraction and segmentation steps. The preprocessed data from these steps form the basis for the subsequent classification and recognition phases. This paper proposes an approach for handwritten Chinese character recognition and classification using only an image alignment technique and does not require the aforementioned steps. Rather than extracting features from the image, which often means building models from very large training data, the proposed method instead uses the mean image transformations as a basis for model building. The use of an image-only model means that no subjective tuning of the feature extraction is required. In addition by employing a fuzzy-entropy-based metric, the work also entails improved ability to model different types of uncertainty. The classifier is a simple distance-based nearest neighbour classification system based on template matching. The approach is applied to a publicly available real-world database of handwritten Chinese characters and demonstrates that it can achieve high classification accuracy and is robust in the presence of noise.

Hierarchical attributed graph representation and recognition of handwritten chinese characters

The theory of traditional Chinese medicine (TCM) is an important part of Chinese culture. In the long history, there are a large number of excellent prescriptions, whose laws have been explored by many studies, but few works directly studied the generation of prescriptions. With the rapid development of deep learning, many applications of text generation using neural networks have emerged. Prescriptions are the doctors’ clinical experience and the results of neural networks also come from the accumulated experience. So, it is very feasible to apply deep learning techniques to the recommendation on TCM prescriptions. GAN and its variants have been applied in text generation recently. It has advantages in many aspects, such as the rapid speed of computation, the update of parameters by back propagation without Markov chain and more real data generation with two-players game. We attempted to know the important attributes of prescriptions and use these contents as the training data for variants of GAN to generate prescriptions. Specifically, we attempted to apply SeqGAN (Sequence Generative Adversarial Nets) and CGAN (Conditional Generative Adversarial Nets) to prescription generations. By underlying the knowledge of TCM, the prescriptions with different characteristics can be successfully generated. In the experiments, we conducted the comparative evaluations on the original data with other models. The results showed that applications in the innovation of prescription sequence generations have certain feasibility and significance, even can provide some reference values for the innovation of TCM.

Recognizing online handwritten Chinese characters using RNNs with new computing architectures

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.

This paper presents a methodology for off-line handwritten Chinese character recognition based on mergence of consecutive segments of adaptive duration. The handwritten Chinese character string is partitioned into a sequence of consecutive segments, which are combined to implement dissimilarity evaluation within a sliding window whose durations are determined adaptively by the integration of shapes and context of evaluations. The average stroke width is estimated for the handwritten Chinese character string, and a set of candidate character segmentation boundaries is found by using the integration of pixel and stroke features. The final decisions on segmentation and recognition are made under minimal arithmetical mean dissimilarities. Experiments proved that the proposed approach of adaptive duration outperforms the method of fixed duration, and is very effective for the recognition of overlapped, broken, touched, loosely configured Chinese characters.

In this paper, the off-line Chinese character image is transformed into ellipse shape of basic Chinese characters strokes in different position. The stroke "turning" and joint or crossover of strokes is combined by basic strokes. The neural network for extracting Chinese character has been build up. The research on off-line Chinese character image is transformed into the research on double weights elliptical neural network based on biomimetic pattern. By extracting fault-tolerant features of the 4 kinds of basic stroke, 13 kinds of "turning" stroke and 7 kinds of consistency and intersection of strokes, the data-knowledge table of features is constructed. The method of bionic recognition gives the machine the ability of cognizing features of stroke type and number, features of stroke location and features of topology structural. The simple and complex handwritten Chinese characters are used to carry out simulation experiment in SCUT-IRAC-HCCLIB. The experiment results show that the algorithm exhibits a strong ability of cognizing handwritten Chinese characters.

Diversified text-to-image generation via deep mutual information estimation

The Hough transform (HT) is a commonly used technique for the identification of straight lines in an image. The Hough transform can be equivalently computed using the Radon transform (RT), by performing line detection in the frequency domain through use of central-slice theorem. In this research, an advanced Radon transform is developed using a multilayer fractional Fourier transform, a Cartesian-to-polar mapping, and 1-D inverse Fourier transforms, followed by peak detection in the sinogram. The multilayer fractional Fourier transform achieves a more accurate sampling in the frequency domain, and requires no zero padding at the stage of Cartesian-to-polar coordinate mapping. Our experiments were conducted on mix-shape images, noisy images, mixed-thickness lines and a large data set consisting of 751,000 handwritten Chinese characters. The experimental results have shown that our proposed method outperforms all known representative line detection methods based on the standard Hough transform or the Fourier transform.

Applied Biomimetic Pattern Recognition to replace “differentiation” of characteristic sample by machine “cognition”, a novel method of handwritten Chinese characters recognition is presented. A double weights elliptical neuron is used to cover four basic kinds of handwritten Chinese characters stroke segment. The topological property among the stroke segment neurons is analyzed. Nine style of Chinese characters stroke with fault tolerance are combined. Imitated typing methods of human Chinese characters font code, the style and number of stroke neurons which have redundant fault tolerant shapes are counted. A kind of characteristic knowledge data-base table of handwritten Chinese characters and the sample data-base of standard printed Chinese characters and fault tolerant matching rules are built. Simple and more complex handwritten Chinese characters in SCUT-IRAC database are tested. The method is proved to be close to human fault tolerance and veracity.

A unique stroke ordering for handwritten Chinese characters is desirable in many applications including efficient character recognition and automatic radical extraction. Although there are some rules or conventions for writing Chinese characters, yet no consistent and complete rule set is available. Besides special radical knowledge is often needed. It is the purpose of this paper to propose a set of rules for stroke ordering for producing a unique stroke sequence for Chinese characters. It requires no special radical knowledge or knowledge of character block layouts, so it is easy for machine implementation. Moreover, the stroke sequences derived are similar to those given in the dictionary, if not the same. To deal with the writing, variations among writers, we generalize the derived stroke structure sequence to obtain a more consistent stroke information. This generalized stroke structural sequence can be used in the handwritten Chinese character preclassification. Experiments showing applications of our method are reported.