Character Recognition Method Research Articles

Implementation of Optical Character Recognition using Tesseract with the Javanese Script Target in Android Application

Recognising characters from text have been a popular topic in the computer vision area. The application can benefit to many problems in the world. For example: recognising text in documents, classifying the text or scripts of documents, plate recognition, etc. Many researchers have been developed the methods for recognising characters in by using Optical Character Recognition methods. Although text recognition problem using Optical Character Recognition has been more or less solved, most of the Optical Character Recognition problem explored is belong to Latin alphabet texts. Meanwhile, there are several languages have non-Latin scripts as the written text. Recognising a non-Latin script is quite challenging as the contour and shape of the text are relatively different with a Latin script text. This research aims to collect datasets for OCR in Javanese characters. A total of 5880 characters were collected and trained with several methods with Tesseract OCR tools. The models then be implemented to a mobile phone (Android based). The highest accuracy (97,50%) achieved by the model was achieved by combining single boundary box for the whole parts of the character and the separate boundary boxes in main body and sandangan parts.

Development of a system for graphic captcha systems recognition using competing cellular automata

Peculiarities of the use of competing cellular automata for problems of recognition of complex captcha systems have been explored. For this purpose, the concept of competing cellular automata has been introduced and a mathematical model of their functioning and interaction has been developed. The mathematical model of competing cellular automata based on the set theory has been described to specify moving cellular automata, which shift to the neighboring states of characters and implement their transition rules in such a way. Based on this mathematical model, a recognition system for captcha images implemented in the code by means of JavaFX 2.0 technology has been developed, which allowed reaching the crossplatformness and correct functioning on different operating systems. The libraries of cellular automata have been developed for the English language. Each symbol of the alphabet is represented in the form of a state system, which is aligned with a cellular automaton with states describing the given symbol. We used Java programming language for development and OpenCV library for the ability to handle images which allowed us to achieve high-quality recognition results. The architecture of the developed system of recognition of complex captcha images in the form of diagrams of classes of the main blocks with detailed descriptions of each class has been considered. Computer experiments have been carried out with different sets of distorted characters used in actual captcha systems and recognition quality indices of the developed software obtained. It has been shown that the probability of obtaining the correct result of captcha image recognition exceeds 80 % with a degree of deformation of characters up to 20 %. With a degree of deformation of characters over 30 %, there is a high probability of false character recognition. The advantages of the method of text character recognition based on competing cellular automata include simplicity of rules of engagement, ability to parallelize the process of recognition easily, capability of recognition of distorted and partially overlapping characters that are the basis of modern captcha systems

A Novel Similar Character Discrimination Method for Online Handwritten Urdu Character Recognition in Half Forms

Online handwritten Urdu character recognition is one of the key technologies for intelligent interface on smart phones and touch screens. It is a challenging research topic as Urdu script has many similar character groups. A novel similar character discrimination method for online handwritten Urdu character recognition is proposed in this paper which includes pre-classification, feature extraction and fine classification process. The pre-classifier enables the discrimination of similar characters by putting them in distinct smaller subsets according to stroke number and diacritics. Then structural features and wavelet features are extracted. Finally, Support Vector Machines (SVM), Artificial Neural Networks (ANN), and Recurrent Neural Network (RNN) classifiers are compared for fine classification within subsets. Results of RNN classifier without using the proposed pre-classifier and features have also been obtained to check the end-to-end capability of the RNN classifier. Experimental results show that the proposed method is efficient and achieves an overall accuracy of 96% on a large-scale self-collected dataset. It is feasible to extend this method for other Arabic scripts.

Development of English-Indonesia Idiom Mobile Dictionary Using Optical Character Recognition

Idioms are phrases that often have different meanings with their literal meaning. The combination of words used, will have different meaning if translated one by one, especially for English words. Currently there are many idiom dictionary applications that exist, but the users have to type the words as the input manually. This requires more precision and time, especially for users who are less fluent in English. To speed up the search process and reduce errors that may occur, optical character recognition methods are used in this study. Through testing and observation to the user, it can be concluded that this application is able to help users to find translations of idioms in English more quickly and precisely.

A Method on The Translation of The Words Written by Ottoman Alphabet to Turkish Alphabet

The Ottoman alphabet was used as a writing language in Turkish with Arabic letters. The Turkish alphabet is used as a writing language in Turkish with Latin letters. There are numerous documents written in the Ottoman alphabet in the archives. In this study, the image of the words written in the Ottoman alphabet was converted into editable text by optical character recognition method. In this way the words are translated into text. Later, the characters in these words were made understandable by using their equivalents in the Turkish alphabet. It was tried to increase the accuracy rate by comparing the words translated into Turkish alphabet and the table where Turkish words were found. An algorithm that gives a similarity value is used for the comparison process.

Robust Character Segmentation and Recognition Schemes for Multilingual Indian Document Images

ABSTRACTThis paper presents robust methods for character segmentation and recognition for multilingual (Latin and Devanagari) Indian document images. The documents degraded over the years because of text diffusion, layout structures, and intermixed texts (printed and handwritten). In character segmentation algorithm, initially, pre-processing like noise and non-uniform illumination reduction is performed. Segmentation columns between characters are obtained using their structural properties. In addition, graph distance theory is used to separate touched and overlapped characters. Finally, support vector machine classifier is employed to validate the segmentation results. For character recognition, set of features based on geometrical properties of the characters are proposed. First feature is formed using center pixel of each non-overlapping block, whereas second feature is a vector of cut points calculated in the triangular areas of the character. Apart from these features, one more feature is used that contains neighborhood information of text pixels. Comprehensive experiments are performed on publicly available databases along with the database collected by us. Benchmarking analysis shows that highest segmentation and recognition accuracies are obtained 98.79% and 99.6%, respectively.

Generative Adversarial Networks for Open Set Historical Chinese Character Recognition

Historical Chinese character recognition has been suffering from the problem of samples labeling, not only the problem of lacking sufficient labeled training samples, but also of sample classes. So the scenario for Historical Chinese character recognition is "open set" recognition, where incomplete labeling of sample classes is present at training time, and unknown classes can be submitted to the system during testing. This paper proposes a method for open set Historical Chinese Character Recognition. For open set recognition, the features available in the training data cannot effectively characterize different kinds of unknown classes. We assume that the features which characterize unknown classes can be derived or learned from other similar data sets. We utilize an auxiliary data set combined with the open set training data set to learn good features to represent historical Chinese characters. The auxiliary data set is translated using Generative Adversarial Networks (GAN) to make sure that the translated data set is as close to the historical Chinese character dataset as possible. Then we construct a neural network for features extraction. The neural network is trained using an alternative training method with the translated auxiliary dataset and incomplete labeled historical Chinese character data set. Last, features are extracted from certain layer of the trained neural network. Unknown samples are detected using statistical modelling of the Euclidean metric between samples. Experimental results show that the proposed method is effective.

Optimization of Convolutional Neural Network for Recognition of Vehicle Frame Number

With the development of the economy and the surge in car ownership, the sale of used cars has been welcomed by more and more people, and the information of the vehicle condition is the focus information of them. The frame number is a unique number used in the vehicle, and by identifying it can quickly find out the vehicle models and manufacturers. The traditional character recognition method has the problem of complex feature extraction, and the convolutional neural network has unique advantages in processing two-dimensional images. This paper analyzed the key techniques of convolutional neural networks compared with traditional neural networks, and proposed improved methods for key technologies, thus increasing the recognition of characters and applying them to the recognition of frame number characters.

Recognition of handwritten Lanna Dhamma characters using a set of optimally designed moment features

Lanna Dhamma alphabet was used mainly for religious communication in the ancient Lanna Kingdom of Thailand. The old manuscripts using this alphabet are gradually decayed. It is desirable to preserve these valuable manuscripts in machine-encoded text files. Existing works used optical character recognition (OCR) methods based on wavelet transform for recognition of handwritten Lanna Dhamma characters. However, the test accuracy of writer-independent recognition is not satisfactory. This work proposes an OCR method, called LDIMS, for recognition of handwritten Lanna Dhamma characters using a set of optimally designed moment features. The LDIMS using an optimization approach to feature selection consists of three main phases: (1) determination of moment orders for each of eight effective moment descriptors, (2) the best combination of selected moment descriptors and (3) the optimized selection of moment features using an inheritable bi-objective genetic algorithm. The LDIMS has three individual feature sets for the recognition of handwritten Lanna Dhamma characters in upper, middle and lower levels. The character images gleaned from previous work were used as a training dataset. A new character image dataset from different writers was established for evaluating ability of writer-independent recognition. The experimental results show that the LDIMS using four moment descriptors, Meixner, Charlier, Tchebichef and Hahn, has test accuracies of 86.60, 74.38 and 85.82% for the characters in upper, middle and lower levels, respectively. The LDIMS with a mean accuracy of 82.27% performed well in recognizing the handwritten Lanna Dhamma characters from new writers, compared to existing methods using generic descriptors in terms of both accuracy and feature number used. Experimental results show that the generalized OCR method, LDIMS, is also effective for character recognition of digit and English alphabets, compared to existing methods.

A REVIEW ON RECOGNITION OF HANDWRITTEN URDU CHARACTERS USING NEURAL NETWORKS

Character recognition being one of the most interesting and attractive areas of pattern recognition and artificial intelligence has got additional consideration during last decade due to its wide range of applications. It contributes immensely to the computerization process and enhancing the man-machine interaction in many applications. It is an art of detecting and recognizing the characters from input image and converting them into ASCII or other corresponding machine editable form. There are four main phases of Character Recognition – Data acquisition and Preprocessing, Segmentation, Feature extraction and Classification. Several research studies have been carried out for recognition of scripts like Chinese, Japanese, English, Devanagari, etc. but the research regarding Urdu Script is still immature due to cursive, variable and overlapping nature of Urdu characters and different writing styles. Research studies on printed Urdu characters have shown good recognition rate but the Handwritten Urdu Script Recognition is still an open and challenging area for researchers. This paper presents a review of Urdu handwritten character recognition methods with special reference to neural networks and includes information regarding the various operations that may be performed on the image for the recognition of Urdu characters. In literature, it has been found that B-Spline curves are not yet applied in combination with Neural Networks for Urdu script recognition. The current research work intends to use B-Splines curves for feature extraction with Feed Forward Neural Network as classifier and focuses on isolated characters in offline domain.

Multi-order co-occurrence activations encoded with Fisher Vector for scene character recognition

Scene character recognition remains a challenging task due to many interference factors. Considering that characters are composed of a series of parts arranged in certain structures, in this paper, we propose a novel representation termed multi-order co-occurrence activations (MCA) encoded with Fisher Vector (FV), namely MCA-FV. It implicitly models the co-occurrence information of discriminative character parts at different orders to boost the recognition performance. We first extract convolutional activations as local descriptors of character parts from convolutional neural networks (CNNs). Then, we introduce MCA features to capture the multi-order co-occurrence cues among different discriminative character parts. Finally, we apply FV to encode co-occurrence features of each order and obtain a global representation of MCA-FV. The proposed method is evaluated on four scene character datasets including English and Chinese datasets. Experiment results demonstrate the effectiveness of the proposed method for scene character recognition.

Fisher vector for scene character recognition: A comprehensive evaluation

Fisher vector (FV), which could be seen as a bag of visual words (BOW) that encodes not only word counts but also higher-order statistics, works well with linear classifiers and has shown promising performance for image categorization. For character recognition, although standard BOW has been applied, the results are still not satisfactory. In this paper, we apply Fisher vector derived from Gaussian Mixture Models (GMM) based visual vocabularies on character recognition and integrate spatial information as well. We give a comprehensive evaluation of Fisher vector with linear classifier on a series of challenging English and digits character recognition datasets, including both the handwritten and scene character recognition ones. Moreover, we also collect two Chinese scene character recognition datasets to evaluate the suitability of Fisher vector to represent Chinese characters. Through extensive experiments we make three contributions: (1) we demonstrate that FV with linear classifier could outperform most of the state-of-the-art methods for character recognition, even the CNN based ones and the superiority is more obvious when training samples are insufficient to train the networks; (2) we show that additional spatial information is very useful for character representation, especially for Chinese ones, which have more complex structures; and (3) the results also imply the potential of FV to represent new unseen categories, which is quite inspiring since it is quite difficult to collect enough training samples for large-category Chinese scene characters.

A New Database of Digits Extracted from Coins with Hard-to-Segment Foreground for Optical Character Recognition Evaluation

Since the release date struck on a coin is important information of its monetary type, recognition of extracted digits may assist in identification of monetary types. However, digit images extracted from coins are challenging for conventional optical character recognition (OCR) methods because the foreground of such digits has very often the same color as their background. In addition, other noises, including the wear of coin metal, make it more difficult to obtain a correct segmentation of the character shape. To address those challenges, this paper presents the CoinNUMS database for automatic digit recognition. The database CoinNUMS, containing 3006 digit images, is divided into three subsets. The first subset CoinNUMS_geni consists of 606 digit images manually cropped from high-resolution photos of well-conserved coins from GENI coin photos; the second subset CoinNUMS_pcgs_a consists of 1200 digit images automatically extracted from a subset of the USA_Grading numismatic database containing coins in different quality; the last subset CoinNUMS_pcgs_m consists of 1200 digit images manually extracted from the same coin photos as CoinNUMS_pcgs_a. In CoinNUMS_pcgs_a and CoinNUMS_pcgs_m, the digit images are extracted from the release date. In CoinNUMS_geni, the digit images can come from the cropped date, the face value or any other legends containing digits in the coin. To show the difficulty of these databases, we have tested recognition algorithms of the literature. The database and the results of the tested algorithms will be freely available on a dedicated website .

A vision-based character inspection system for tire mold

PurposeThis paper aims to design an automatic inspection system for the characters on tire molds, which involves a vision-based inspection method for the characters on tire molds.Design/methodology/approachAn automatic inspection equipment is designed according to the features of the tire mold. To implement the inspection task, the corresponding image processing methods are designed, including image preprocessing, image mosaic, image locating and character inspection. Image preprocessing mainly contains fitting the contours of the acquired tire mold images and those of the computer aided design (CAD) as the arcs of two circles and polar transformation of the acquired images and the CAD. Then, the authors propose a novel framework to locate the acquired images into the corresponding mosaicked tire mold image. Finally, a character inspection scheme is proposed by combining an support-vector-machine-based character recognition method and a string matching approach. At the stages of image locating and character inspection, image mosaic is simultaneously used to label the defects in the mosaicked tire mold image, which is based on histograms-of-gradients features.FindingsThe experimental results indicate that the designed automatic inspection system can inspect the characters on the tire mold with a high accuracy at a reasonable time consumption.Practical implicationsThe designed automatic inspection system can detect the carving faults for the characters on the tire molds, which are the cases that the characters are wrongly added, deleted or modified on the tire mold.Originality/valueTo the best of the authors’ knowledge, this is the first automatic vision-based inspection system for the characters on tire molds. An inspection equipment is designed and many novel image processing methods are proposed to implement the inspection task. The designed system can be widely applied in the industry.

Research on a Method of Character Recognition for Self-learning Errors

Research on a Method of Character Recognition for Self-learning Errors

一种多样变换的手写验证码自动识别算法的研究及应用

研究验证码自动识别技术可以进一步提升人识别验证码的可读性，增强机器识别的难度，从而提高网络安全性。针对目前提出的验证码识别方法基本都是采用光学字符识别(OCR)方法对机器写的标准字符进行识别，本文提出了一种多样变换的手写验证码自动识别算法，对彩色验证码进行识别主要包括彩色验证码的二值化、手写字符的区域分割、同一字符的区域连接、使用卷积神经网络对手写字符进行训练、手写字符识别。本文的实现结果明显优于OCR的识别结果。结果表明通过该网站的测试，基本上能自动识别该网站的验证码。 Research on technique of automatically identifying CAPTCHA can promote people to identify the readability of verifying the code further and strengthen the difficulty that the machine identifies and raises a network safety thus. Currently aiming at the CAPTCHA identification methods are basically used optical character recognition (OCR) method to identify the standard characters written by the machine. The paper puts forward color CAPTCHA identifies to mainly include color verification code binary by threshold, the connect district segmentation of the handwritten character list, the near district of the same character list links, use convolution neural network to train a character and to identify handwritten character. The paper realization obviously surpasses an identifying of OCR result. The result shows CAPTCHA of the website that is basically passed by the website test; the website can automatically identify CAPTCHA of the website.

Handwritten character recognition technology combined with artificial intelligence

The offline handwritten character recognition is an important branch of character recognition, which involves character recognition, image processing, digital signal processing, artificial intelligence, fuzzy mathematics, information theory, computer and other disciplines. In the process of off-line handwritten character recognition, it only processed two-dimensional character dot images, and there are problems such as too many character classes, complex font structure, large deformation of handwritten characters and other issues. Currently off-line handwritten character recognition technology is still immature, which is still in the laboratory research stage. This paper attempted to combine neural network technology and character recognition technology, and came up with effectively new method of handwritten character recognition, finding a feasible new way to solve practical difficulties between handwritten character recognition. In this paper, according to the building process of the multi-level recognition system, the article introduced the preprocessing of character image, character feature extraction, neural networks and optimization algorithms as well as multiple neural networks in turn. In the end, based on experimental results by identification system in Matlab, the author comparatively discussed the performance of the recognition system, analyzing the notices of establishing the identification system.

Manchu character segmentation and recognition method

Discusses the text segmentation of Manchu character recognition process, improved the traditional projection segmentation method, improves the segmentation accuracy of Manchu. The nearest neighbor and support vector machine, Bayes three Manchu recognition method, combined with the characteristics of Manchu, a design based on the printed Manchu recognition system of multi classifier fusion decision, realize image preprocessing, Manchu character recognition and other functions. The experimental results show that, the system has achieved good recognition effect of Manchu.

Topology-Based Character Recognition Method for Coin Date Detection

For recognizing coins, the graved release date is important information to identify precisely its monetary type. However, reading characters in coins meets much more obstacles than traditional character recognition tasks in the other fields, such as reading scanned documents or license plates. To address this challenging issue in a numismatic context, we propose a training-free approach dedicated to detection and recognition of the release date of the coin. In the first step, the date zone is detected by comparing histogram features; in the second step, a topology-based algorithm is introduced to recognize coin numbers with various font types represented by binary gradient map. Our method obtained a recognition rate of 92% on synthetic data and of 44% on real noised data.

A holistic approach for Off-line handwritten cursive word recognition using directional feature based on Arnold transform

This paper presents a holistic off-line handwriting recognition system based on extraction of directional features which depends on the stroke orientation distribution of cursive word. This stroke orientation distribution is estimated using Arnold transform followed by Hough transform. Besides this feature some other directional shape features are also used to form feature vector. Finally, a multi-class linear SVM is employed to recognize cursive word. Experiments are carried out on CENPARMI database of legal amount written in English and an overall accuracy of 87.19% is achieved. We have also compared our proposed method with the state-of-the-art methods for handwritten character recognition using C-Cube data-set.