Search

One of the algorithms stored in natural intelligence is the writing of Arabic numerals in Indonesian words. Algorithms in naturals intelligence are not easy to find. This problem gave us an idea to create artificial intelligence that tries to mimic natural intelligence algorithms. The proposed algorithm for building artificial intelligence is an R-Z rule-based system. This rule-based system contains a knowledge base of R-Z rules and a knowledge base of facts. In the knowledge base, the R-Z rule provides the R rule and the Z rule, while the facts knowledge base provides facts in the form of a definite standard number and an affix word. R-Z rule-based system for reasoning writing Arabic numerals in Indonesian words uses forward chaining. Artificial intelligence designs that mimic naturals intelligence in writing numbers in Indonesian words were made in C using Borland C++ 5.02 software. The experimental results show that by applying the R's rule of seven rules and Z's of twenty-five rules, the R-Z rule-based system can write Arabic numerals in Indonesian words from Arabic numerals "0" to Arabic numerals "9999999". For example, to write the Arabic number "10" in Indonesian words, the R-Z rule-based system starts with the R2 rule. Rule R2 takes action on Z3 to create new facts about Arabic numerals in the Indonesian word, namely "SEPULUH."

Using Generative Adversarial Networks (GANs) have led to significant improvements in computer vision research filed. GANs are used to generate novel images by pitting two neural networks against each other, where a generator network generates images from a latent space and a discriminator network classifies the generated images. In our work, a number of different GANs is applied to generate handwritten Arabic digits in the Eastern Arabic numeral script and determine which GAN flavor preforms best. To generate these digits, we test the following types of GANs: a Basic GAN, Deep Convolutional GAN (DCGAN), Bidirectional GAN (BiGAN), VanillaGAN, and WassersteinGAN (WGAN). Then, a Frechet Inception Distance (FID) measurement tool along with native-Arabic human evaluation are carried out to evaluate the generated images. Experimental results show that DCGAN outperforms other employed GANs with respect to FID benchmark with a value of 96.815%. While DCGAN outperforms other GANs on FID, BiGAN obtains best performance with respect to native-Arabic human benchmark with a value of 69.93%. Our promising experimental results illustrate the ability of using GAN models to develop huge handwritten Arabic datasets with less efforts comparing with traditional methods of collecting data.

Melisa Vivanco objects to my theory of the Arabic numerals in Roads to Reference that the reference fixing procedure that I postulate doesn’t exploit the morphological structure of the Arabic numerals, but it should. Against Vivanco, I argue that the procedure in question does exploit the morphological structure of the numerals in an essential way.

Children’s spontaneous focusing on Arabic number symbols (SFONS) has been identified as a relevant component of their early mathematical development. This study investigated whether SFONS is a sepa...

Here we report the results of a speeded relative quantity task with Chinese participants. On each trial a single numeral (the probe) was presented and the instructions were to respond as to whether it signified a quantity less than or greater than five (the standard). In separate blocks of trials, the numerals were presented either in Mandarin or in Arabic number formats. In addition to the standard influence of numerical distance, a significant predictor of performance was the degree of physical similarity between the probe and the standard as depicted in Mandarin. Additionally, competing effects of physical similarity, defined in terms of the Arabic number format, were also found. Critically the size of these different effects of physical similarity varied systematically across individuals such that larger effects of one compensated for smaller effects of the other. It is argued that the data favor accounts of processing that assume that different number formats access different format-specific representations of quantities. Moreover, for Chinese participants the default is to translate numerals into a Mandarin format prior to accessing quantity information. The efficacy of this translation process is itself influenced by a competing tendency to carry out a translation into Arabic format.

Understanding of arithmetic students in Arabic learning is still low. This study aims to determine the effectiveness of the Direct Method in learning Arabic number material in Arabic understanding Arithmetic Students. The approach used is experimental research approach with Posttest-Only Control Design. The population and sample of the study were determined through the Random Sampling technique. Analysis of the data used included normality test, homogeneity test, average similarity test, completeness test, and comparative test. (1) the arithmetic understanding of students taught by the Direct Method in learning ‘Adad dan Ma’dud material can exceed the Standard of Minimum Mastery (KKM) of 76,87 , (2) the arithmetic understanding of students taught by the Direct Method is 76,87 higher than from the conventional models 65,32 . Then it can be concluded that the Direct Method is effective in understanding arithmetic in learning ‘Adad dan Ma’dud material. The novelty of this research is that using the Direct Method can facilitate a direct understanding of 'adad and ma'dud. This research contributes in Arabic teaching by using the Direct Method in the material 'Adad and Ma'dud to improve students' arithmetic understanding .

Numeral recognition is an important preliminary step for optical character recognition, document understanding and others. Deep learning based numeral recognition models have achieved huge popularity among the researchers around the world since last few years. Several convolutional neural network (CNN) based architectures have been proposed and many of those have achieved state-of-the-art results in numeral recognition. In this paper, we have explored CNN based architectures for handwritten Arabic numeral recognition. We have also developed a handwritten Arabic numerals dataset using various morphological operations on an existing dataset thus increasing the size of the dataset from 3000 to 72,000 images. A modification of previously proposed CNN architecture has given us an accuracy of 98.91% and our proposed architecture has produced 99.76%, which is comparable to state-of-the-art results found in the domain of handwritten Arabic numeral recognition.

Chinese characters are an important medium for Chinese people to exchange information and perceive the world. With the advent of the information age, a large number of paper documents need to be electronically stored and shared. The recognition accuracy of paper printed Chinese characters and online handwritten Chinese characters has reached a high level. However, offline handwritten Chinese characters have a variety of styles and shapes. There is still a lot of room for improvement in the current recognition accuracy due to a large number of similar characters. This work proposes a "private customized" handwritten Chinese character recognition system based on convolutional neural networks. Users can train their neural network model according to their own writing style. The system includes four parts: character segmentation, Chinese character labeling, neural network training, and predictive recognition. The system can independently expand the data set during use, thereby continuously improving the recognition accuracy. The experimental results show that this method has a good recognition effect for Chinese characters, English letters, Arabic numerals, and punctuation marks, and the recognition accuracy rate can reach more than 98%.

In the field of continuous hand-gesture trajectory recognition, aiming at the problems of existing a lot of noise for handwriting trajectories, and difficult to segment multiple continuous hand gestures accurately, a long short-term memory-based dynamic probability (DP-LSTM) method is proposed. Firstly, obtain the classification result for each sub-period in the whole time period by using LSTM; secondly, cluster the classification results by non-maximum suppression for trajectory algorithm to eliminate interference of invalid subsets; Finally, the end point of the valid trajectory is obtained according to the characteristics of the probability change, thus realising dynamic trajectory segmentation and recognition. In order to evaluate the performance of the DP-LSTM, this method is evaluated by using an Arabic numerals gesture database. The experiments show that the DP-LSTM has a high recognition rate for continuous hand gestures and can recognise its in real time.

Latencies and amplitudes of the Р100, N170 and P200 components of event-related potentials (ERPs) derived during visual presentation of Roman and Arabic numerals from healthy subjects and in patients with paranoid schizophrenia were studied. The activation level was estimated as the ratio of the values in response to Arabic and Roman numerals. The Р100 latency in response to Arabic numerals in controls was shorter in the occipital and parietal areas, while the amplitude was higher in the left posterior temporal area. In schizophrenia patients, the Р100 amplitude in response to Arabic numerals was higher in the left temporal area and lower in the right central area. The intergroup analysis showed that the Р100 amplitude in the central parietal, left temporal and right posterior temporal areas in response to Arabic numerals in the control was greater than in schizophrenia patients. The N170 latency in response to Arabic numerals in the control group was shorter in the left occipital, right temporal and parietal areas, while the amplitude was lower in the right posterior temporal area. In schizophrenia patients, the N170 latency of response to Arabic numerals was shorter only in the middle parietal area. Intergroup analysis showed that the N170 latency of responses to both types of stimuli was shorter in the occipital areas in controls compared to the patients, and the amplitude of this wave in response to Roman numerals was higher in the right posterior temporal area. The P200 latency in response to Arabic numerals in control group was shorter in the left frontal area, while the amplitude was higher in the middle frontal and left temporal areas. In schizophrenia patients, the P200 latency in response to Arabic numerals was longer in the left occipital region and the amplitude was lower in the middle central and right occipital areas. Intergroup analysis showed that the latency in response to Arabic numerals was shorter in the left parietal area in the control group compared to the patients, and the amplitude in the right central and left temporal areas was higher. The P200 amplitude in response to Roman numerals in the right prefrontal and left frontal areas was higher in the control compared to the patients. Thus, in norm and in schizophrenia patients different levels of cortical areas activation to the two types of stimuli were revealed reflecting the differences of these stimuli characteristics.

Hand gestures are spatio-temporal patterns which can be characterized by collections of spatio-temporal features. Recognition of hand gestures is to find the re-occurrences of such spatio-temporal patterns through pattern matching. However, dynamic hand gestures have many obstacles for accurate recognition, including poor lighting conditions, camera’s inability to capture dynamic gesture in focus, occlusion due to finger movement, color variations due to lighting conditions. The Microsoft Kinect device provides an effective way to solve the above issues and also provides the skeleton for more convenient hand localization and tracking. The aim of this study is to develop a new trajectory-based method for hand gesture recognition using Kinect. In the first step, trajectory-based hand gesture features including spatial position and direction of fingertips, are derived from Kinect. The properties associated with the hand motion dynamics are preserved in these features. In the second step, radial basis function (RBF) neural networks are employed to model and approximate the hand motion dynamics derived from different hand gestures which represent Arabic numbers (0–9) and English alphabets (A–Z). The trained patterns of the approximated hand motion dynamics is stored in constant RBF networks. In the last step, a bank of dynamical estimators is constructed for all the training patterns, in which the constant RBF networks are embedded in. By comparing the set of estimators with a test gesture pattern, a set of recognition errors are generated, in which the average $$L_1$$ norms of the errors are taken as the recognition measure based on the smallest error principle. Finally, experiments are carried out to assess the performance of the proposed method compared with other state-of-the-art approaches. By using the twofold and tenfold cross-validation styles, the correct recognition rates for Arabic numbers (0–9) and English alphabets (A–Z) are reported to be $$95.83\%, 97.25\%$$ , and $$91.35\%, 92.63\%$$ , respectively.

Arabic and Hindi handwritten numeral detection and classification is one of the most popular fields in the automation research. It has many applications in different fields. Automatic detection and automatic classification of handwritten numerals have persistently received attention from researchers around the world due to the robotic revolution in the past decades. Therefore, many great efforts and contributions have been made to provide highly accurate detection and classification methodologies with high performance. In this paper, we propose a two-stage methodology for the detection and classification of Arabic and Hindi handwritten numerals. The classification was based on convolutional neural networks (CNNs). The first stage of the methodology is the detection of the input numeral to be either Arabic or Hindi. The second stage is to detect the input numeral according to the language it came from. The simulation results show very high performance, the recognition rate was close to 100%.

This study focuses on senior citizens as the target group for font design. It sets the field in the visual display terminal (VDT) operating environment. By referring to the existing literature and research results, we not only find the current fonts with better legibility but also understand the visual problems faced by senior citizens and what is suitable for In addition to finding the existing fonts with better legibility, we know the visual difficulties faced by the elderly and what is the appropriate graphic design for the elderly and explore the relevant criteria of font design in depth. Then, through experiments to find similar numerical combinations and computer simulations of older people’s reading conditions, the strengths and weaknesses of existing fonts in mild and severe cataract environments were identified, and the elements of font design were analyzed and investigated. Finally, we hope to design an Arabic numeral font suitable for senior citizens and evaluate it through an actual operation.

Handwritten signatures widely exist in our daily lives. The main challenge of signal recognition on handwriting is in the development of approaches to obtain information effectively. External mechanical signals can be easily detected by triboelectric nanogenerators which can provide immediate opportunities for building new types of active sensors capable of recording handwritten signals. In this work, we report an intelligent human-machine interaction interface based on a triboelectric nanogenerator. Using the horizontal-vertical symmetrical electrode array, the handwritten triboelectric signal can be recorded without external energy supply. Combined with supervised machine learning methods, it can successfully recognize handwritten English letters, Chinese characters, and Arabic numerals. The principal component analysis algorithm preprocesses the triboelectric signal data to reduce the complexity of the neural network in the machine learning process. Further, it can realize the anticounterfeiting recognition of writing habits by controlling the samples input to the neural network. The results show that the intelligent human-computer interaction interface has broad application prospects in signature security and human-computer interaction.

This paper proposes machine learning and deep learning techniques for recognizing Arabic handwritten text. Finally, this paper introduces a comparative study between them in term of their performance. Actually, the classification of Arabic handwritten text plays a vital role in the computer vision domain, where traditional machine learning techniques and deep learning techniques are commonly used by researchers. In this paper, both machine learning and deep learning techniques are proposed and evaluated for recognizing Arabic handwritten text. Several experiments were carried out using both machine learning and deep learning on two different databases the AHCR and ADBase. The AHCR contains 28000 images of handwritten Arabic alphabet letters written by 100 writers. While the ADBase contains 70,000 images of handwritten Arabic digits written by 700 writers. The experimental results on both databases have demonstrated that the performance of the deep learning outperforms machine learning.

This paper proposes an improved convolutional neural network structure which greatly reduces the scale of network training parameters. It uses the global average pooling algorithm instead of the full connection algorithm to improve the LeNet-5 network. The number of convolution kernels is increased, while the number of subsampling layers is reduced to an optimal value. After verification with MINST handwritten Arabic numeral data set, the results show that the improved network training parameters are only 34.8% of the original, and the recognition accuracy can achieve 99.3%.

The paper presents the implementation of a system for license plate detection and recognition in digital images. A comparative analysis of the detection methods was carried out. It led to the selection of the convolutional neural network for the detection module. The customized network architecture was developed on the basis of the neural network GoogleNet. An overview of approaches for license plate detection and recognition in digital images was made. The best solution for the recognition module was determined. The developed system for license plate detection and recognition can identify license plates that use letters of the Latin alphabet and Arabic digits. Testing of the detection and recognition system of the license plate was carried out. It proved the reliability of the system and its ability to perform well, even in conditions that might prevent the detection and recognition of license plates. The accuracy of the implemented system was at least 94%.

The aim of this study was to examine language switching in a two-digit number naming task. In contrast to single digits, two-digit numbers have a composition rule (i.e., morphological configuration) that may differ between languages. For example, the Arabic number 21 is read with an inverted composition rule in German (unit before decade) and a non-inverted composition rule in English (decade before unit). In the present experiment, one group of German native speakers and one group of Spanish native speakers had to name two-digit numbers in German, English, or Spanish. The results demonstrate a language-switch cost, revealing better performance in language repetition than in language-switch trials. This switch cost was further modulated by repeating or switching the composition rule, since the language repetition benefit (i.e., the switch cost) was reduced in trials with composition-rule switches compared with trials with composition-rule repetitions. This finding indicates that the language in which the number word has to be produced and its composition rule are not switched independently but rather may be integrated into one language schema.

Historical manuscripts and archival documentation are handwritten texts which are the backbone sources for historical inquiry. Recent developments in the digital humanities field and the need for extracting information from the historical documents have fastened the digitization processes. Cutting edge machine learning methods are applied to extract meaning from these documents. Page segmentation (layout analysis), keyword, number and symbol spotting, handwritten text recognition algorithms are tested on historical documents. For most of the languages, these techniques are widely studied and high performance techniques are developed. However, the properties of Arabic scripts (i.e., diacritics, varying script styles, diacritics, and ligatures) create additional problems for these algorithms and, therefore, the number of research is limited. In this research, we first automatically spotted the Arabic numerals from the very first series of population registers of the Ottoman Empire conducted in the mid-nineteenth century and recognized these numbers. They are important because they held information about the number of households, registered individuals and ages of individuals. We applied a red color filter to separate numerals from the document by taking advantage of the structure of the studied registers (numerals are written in red). We first used a CNN-based segmentation method for spotting these numerals. In the second part, we annotated a local Arabic handwritten digit dataset from the spotted numerals by selecting uni-digit ones and tested the Deep Transfer Learning method from large open Arabic handwritten digit datasets for digit recognition. We achieved promising results for recognizing digits in these historical documents., European Research Council (ERC) Project: "Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000" (; UrbanOccupationsOETR); European Union (European Union); Horizon 2020

CAPTCHA (Completely Automated Public Turing test to tell Computer and Human Apart) is widely used as a standard security mechanism to protect resources on websites. Among various kinds of CAPTCHAs, the text-based CAPTCHA is the most popular scheme, which consists of English letters, Arabic digits and other character sets, such as Chinese characters. Due to the large quantity of Chinese characters and complicated character structure, it is difficult for bots to crack Chinese character CAPTCHAs. Thus, Chinese character CAPTCHAs have been widely applied in China. Nevertheless, effective offensive approaches are necessary to help CAPTCHA designers find security vulnerabilities to improve defense mechanisms. To deal with variable-length Chinese character CAPTCHAs with noises, an automatic attacking approach is proposed, which includes preprocessing, character segmentation and character recognition. For character recognition, two methods are proposed: MGLCR (Multi-scale Gabor and Logistic regression based CAPTCHA Recognition) and CCR (Convolutional neural network based CAPTCHA Recognition). MGLCR extracts features by multi-scale Gabor filters and classifies characters with logistic regression. CCR extracts features and recognize characters automatically with CNN (Convolutional Neural Network). Experimental results show that the proposed approaches are efficient in attacking variable-length Chinese character CAPTCHAs with noises. The pros and cons of proposed MGLCR and CCR methods are discussed, which outperform state-of-the-art methods. Besides, the proposed methods could achieve satisfactory results in breaking the mixed character CAPTCHAs which consist of English letters, Arabic digits, Chinese characters and mathematical operators.

The process of classification and recognition of the digits is very important and accurate issue, it should depend on a strong, adequate rule, because there are differences in the representation of the digits for various type of font, rather than to get homogeneous classification for group of digits to show the form or the constant shape for the digit from groups of font , where to be one shape represents each group of the shapes and sizes. This paper proposed a new method for recognition images of Arabic digits with various types and size of font which captured by various kind of camera and various resolution depending on the characteristics of each digit which extracted from study of ten types of font with various size for each type. The characteristic of all shapes and sizes extracted for each digit. Then made a constant model for each digit to be compare with the digit image in any size or font type to recognize the digit in the image. A database of ten templates (one for each digit) built, and these templates are basic and general to recognize the Arabic digits with various font and size. The new method applied for recognition of Arabic digits in the images of license plate of the passenger cars. the recognition percentage was 98%.

A number line consisting of Arabic numerals is a commonly used instructional tool for teaching price comparison. However, typical number lines lack concrete visual cues, which may benefit students with autism spectrum disorders (ASD) who have not yet mastered the representation of Arabic numerals.This study investigated the effects of additional visual cues (i.e., dots) by comparing two types of app-based number line conditions: number lines with and without dots. A single-subject, alternating treatment design study was employed across five secondary students with ASD.Both number line conditions were effective for four of the students in assisting them to select cheaper items and complete task analysis steps. The number line with dots was effective or slightly more effective in selecting smaller numbers for three of the students.The findings of this study support the literature on the use of number lines as an effective tool to assist students in price comparison. The benefits of adding concrete visual cues and other teaching strategies (e.g., the holistic and decomposition models) were discussed. Implications for Rehabilitation This study investigated the effectiveness of concrete visual cues, such as dots, on a number line app for teaching students with ASD who had not yet developed the association of quantities with the numerals. We found that incorporation of a hybrid number comparison model - first holistic (for whole numbers) and then decomposition (for numbers after the decimal point) - is effective when teaching students how to compare prices with an uneven number of digits. This study provides an alternative for special education teachers to schedule practice, such as the use of simulated settings to achieve mastery, then transitioning to community-based settings to test skill generalization.

The present study investigated the influence of lexical links and conceptual mediation in number word translation. In four experiments, a semantic blocking paradigm was used with numerical stimuli and stimuli from other non-number categories. The performance of Spanish L1–English L2 bilinguals was examined in backward and forward translation (Experiment 1 and 3), in naming Arabic digits and pictures (Experiment 2A) and in reading L1 and L2 words (Experiment 2B). The blocked context produced a facilitatory effect for numerical stimuli in the lower range. This result contrasts with the interference effect observed for stimuli from non-number categories in translation and naming tasks. These results are interpreted to mean that the translation of number words, particularly those in the lower range of the numeric sequence, takes place with a large influence of the lexical connections. It supports the suggestions that the influence of processing routes depends on the type of items.

System for converting Arabic numerals to their textual equivalence is an important tool in Natural Language processing (NLP) especially in high-level speech processing and machine translation. Such system is scarcely available for most African languages including the Igbo language. This translation system is essential as Igbo language is one of the three major Nigerian languages feared to be among the endangered African languages. The system was designed using sequence as well as activity diagram and implemented using the python programming language and PyQt. The qualitative evaluation was done by administering questionnaires to selected native Igbo speakers and experts to provide preferred representation of some random numbers. The responses were compared with the output of the system. The result of the qualitative evaluation showed that the system was able to generate correct and accurate representations for Arabic numbers between 1-1000 in Igbo language being the scope of this study. The resulting system can serve as an effective teaching and learning tool of the Igbo language.

It has been shown that when two arrays of Arabic numerals were briefly presented, observers could accurately indicate which array contained the larger number of a target numeral. This study investigated whether this rapid proportion comparison can be extended to other meaningful symbols that share some of notable properties of Arabic numerals. We tested mainly several Japanese Kanji letters, each of which represents a meaning and can work as a word. Using physically identical stimulus sets that could be interpreted as different types of letters, Experiment 1 first confirmed the rapid proportion comparison with Arabic numerals for Japanese participants. Experiment 2 showed that the rapid proportion comparison can be extended to Kanji numerals. Experiment 3 successfully demonstrated that rapid proportion judgments can be found with non-quantitative Kanji letters that are used frequently. Experiment 4 further demonstrated the rapid proportion comparison with frequently used meaningful non-letter symbols (gender icons). The rapid processing cannot be attributed to fluent processing of familiar items, because it was not found with familiar phonograms (Japanese Kana letters). These findings suggest that the rapid proportion comparison can be commonly found with frequently used meaningful symbols, even though their meaning is not relevant to the task.

Thirteenth-century music theory, which followed the ideas of Boethius, was very largely concerned with the numerical proportions associated with musical intervals. Numbers provided an intellectual foundation that did not suffer from the vagaries of the senses. In general neither Boethius nor his greatest exponent, Jacobus (writing c. 1320), explained how they obtained the numbers they used. In this essay I attempt to reconstruct their methods and show how they developed ideas from the first-century Nicomachus to achieve their aims. Jacobus is explicit in saying that the use of the relatively newly introduced methods of algorism – calculating with Arabic numerals – made his cogitations easier. I shall argue that the manuscripts we have of his Speculum musicae show that Jacobus did indeed use algorism in his work.

During the late sixteenth and seventeenth centuries, English men and women replaced their existing oral and object-based arithmetical practices with literate practices based on Arabic numerals. While the adoption of Arabic numerals was incentivized by continental commercial developments, this article argues that England's increasing literacy rates and the development of vernacular arithmetic textbooks enabled changing arithmetical practices. By exploring the qualities of printed books, analyzing marginalia in arithmetic textbooks, and examining changing educational advertisements and curricula over time, we can demonstrate the importance of literacy and literature to early modern arithmetical education.

Individuality of handwriting inserts varying curvatures and angles whenever someone writes a sample of a particular numeral which makes the task of its off-line recognition more challenging. The paper addresses both these issues in novel and robust ways by merging two Digital domains, namely Digital Communications and Digital Image Processing. Curvature is treated by finding analytical features based on distance and slope. Distance based treatment is done by means of Delta Distance Coding whereas slope based analysis is executed with Delta Slope Coding. Angular variations have been countered with the help of rotation invariant physical feature i.e., Pixel Moment of Inertia. A due stress has been laid on Pixel Moment of Inertia by finding it both globally and locally in terms of Centroidal Moment of Inertia and Zonal Moment of Inertia respectively. The above mentioned features are further supported with statistical features in order to differentiate very similar looking numeral pairs like (3, 8), (1, 7), (7, 9). Feature extraction methods are devoid of cumbersome calculations, and classifiers are capable of yielding instantaneous results. Therefore, the current system is a real time system. The system has been tested on unconstrained MNIST dataset. The overall recognition accuracy of 99.26% has been obtained.

Speech recognition has undergone tremendous advancement over the past 50 years. Deep Neural Network (DNN) is one of the most popular methods for speech analysis thanks to its ability to minimize error rate for optimization problems. This research proposes an Arabic digits speech recognition model utilizing Recurrent Neural Network (RNN). The speech recognition model select the finest speech signal representation by feature extraction of Mel-Frequency Cepstrum Coefficients (MFCCs) after having been processed for noise reduction and digits separation. Extracted features from speech of digit are fed into a network with Long Short-Term Memory (LSTM) cells. The LSTM cells have the capability to solve problems associated with temporal dependencies requiring long-term learning and solve the vanishing gradient problems associated with RNN. A dataset of 1040 samples of spoken Arabic digits from different dialects are used in this study where 840 samples used to train the network and another 200 samples are used for testing purpose. The model training is carried out using a computing system with Graphics Processing Unit (GPU). The LSTM model learning parameters is tuned for optimization purpose achieving a higher accuracy of 94% during model training. The testing results of the tuned parameters model shows that the LSTM model can achieve 69% in accuracy when recognizing spoken Arabic digits. The model has the highest accuracy, i.e. 80%, when recognizing the digit zero.

Text effects are combinations of visual elements such as outlines, colors and textures of text, which can dramatically improve its artistry. Although text effects are extensively utilized in the design industry, they are usually created by human experts due to their extreme complexity; this is laborious and not practical for normal users. In recent years, some efforts have been made toward automatic text effect transfer; however, the lack of data limits the capabilities of transfer models. To address this problem, we introduce a new text effects dataset, TE141K, with 141,081 text effect/glyph pairs in total. Our dataset consists of 152 professionally designed text effects rendered on glyphs, including English letters, Chinese characters, and Arabic numerals. To the best of our knowledge, this is the largest dataset for text effect transfer to date. Based on this dataset, we propose a baseline approach called text effect transfer GAN (TET-GAN), which supports the transfer of all 152 styles in one model and can efficiently extend to new styles. Finally, we conduct a comprehensive comparison in which 14 style transfer models are benchmarked. Experimental results demonstrate the superiority of TET-GAN both qualitatively and quantitatively and indicate that our dataset is effective and challenging., Comment: Accepted by TPAMI 2020. Project page: https://daooshee.github.io/TE141K/

One of the challenges in recognizing handwritten texts is the individual style of writing. There is the structural similarity of the different digits to each other in writing. Along with the mentioned challenge, these similarities in the text may increase and make it difficult to correctly recognize digits and numbers. In this paper, a new model for recognizing handwritten digits is presented. The proposed model is a stacking ensemble classifier. This classifier is based on the convolutional neural network (CNN) and the bidirectional long-short term memory (BLSTM). Another innovation of the model is the use of the probability vector of images class as the input of the meta-classifier layer. One of the strengths of BLSTM is the ability to learn arrays and vectors; therefore, from a technical point of view, considering the output probability vector of the first model as the input of the meta-classifier (BLSTM) improves the accuracy of the deep learning model. The reason for using stacking ensemble classification is the sameness of the main body of some Persian/Arabic digits (e.g., “2, 3, and 4”). Also, the style the author writes makes classes that are not similar in a structure similar to each other, which causes errors incorrect recognition. This model helps to recognize the correct set of input digits by examining the structure of similarities. To achieve a reliable result in the face of this challenge, this model has been tested on a large Persian/Arabic dataset to cover a wide range of writing styles from different individuals. The dataset has a total of 102,352 data which 60,000 of them are for training data and 20,000 of them are for test data in ten classes of digits that are used in this paper. The result of using this database is to improve the recognition performance of these challenging digits. In examining the datasetpresented by the model, the accuracy rate of the training set was 99.98%. And the sample accuracy rate in the test set was 99.39%. That is, compared to experimenting with the convolutional neural network and other researches, the rate has increased. (All codes available at http://web.nit.ac.ir/ h.omranpour/ .)

The French clergyman Gerbert of Aurillac, later to become Pope Sylvester II, is said to have represented all numbers with only “nine symbols” on his abacus and is therefore regarded as the one who brought the Arabic numerals to Europe. In his own writings, however, he only used the Roman numerals. Early illustrations of Arabic numerals appear in Europe around the year 1000 in the Abacus treatise of Bernelinus of Paris, who was probably a pupil of Gerbert. In two manuscripts of his “Liber abaci” the number “3” is displayed in a peculiar shape, reminiscent of the sign for “three” in Roman shorthand. Since Gerbert used this shorthand, the sign in question could be an indication of his role in the introduction of the new numbers. In the further course of the study an attempt is made to retrace the possible course of the transmission of Arabic numerals from Muslim Spain to Europe.

In mathematically literate societies, numerical information is represented in 3 distinct codes: a verbal code (i.e., number words); a digital, symbolic code (e.g., Arabic numerals); and an analogical code (i.e., quantities; Dehaene, 1992). To communicate effectively using these numerical codes, our understanding of number must involve an understanding of each representation as well as how they map to other representations. In the current study, we looked at 3- and 4-year-old children’s understanding of Arabic numerals in relation to both quantities and number words. The results suggest that the mapping between quantities and numerals is more difficult than the mapping between numerals and number words and between number words and quantities. Thus, we compared 2 competing models designed to investigate how children represent the meanings of Arabic numbers—whether numerals are mapped directly to the quantities they represent or instead if numerals are mapped to quantities indirectly via a direct mapping to ...

This paper investigates the feed forward back propagation neural network (FFBPNN) and the support vector machine (SVM) for the classification of two Maghrebian dialects: Tunisian and Moroccan. The dialect used by the Moroccan speakers is called "La Darijja" and that of Tunisians is called "Darija". An Automatic Speech Recognition System is implemented in order to identify ten Arabic digits (from zero to nine). The implementation of our present system consists of two phases: The features extraction using a variety of popular hybrid techniques and the classification phase using separately the FFBPNN and the SVM. The experimental results showed that the recognition rates with both approaches have reached 98.3 % with FFBPNN and 97.5 % with SVM.

The main aim of this work is to compare Hindu and Arabic digits with respect to a machine’s ability to recognize them. This comparison is done on the raw representation (images) of the digits and on their features extracted using two feature selection methods. Three learning algorithms with different inductive biases were used in the comparison performed using the raw representation; two of them were also used to compare the digits using their extracted features. All classifiers gave better results for Hindu digits in both cases; when raw representation was used and when the selected features where used. The experiments also show that Hindu digits can be classified with better accuracy, higher confidence and using fewer features than Arabic digits. These results indicate that hand-written Hindu digits are actually easier to recognize than hand-written Arabic digits. The machine learning methods used in this work are instance based learning (the kNN algorithm), Naive Bayesian and neural networks. T...

Due to the technological advances in recent years, paper scientific documents are used less and less. Thus, the trend in the scientific community to use digital documents has increased considerably. Among these documents, there are scientific documents and more specifically mathematics documents.In this context, we present our own dataset of handwritten mathematical symbols composed of 10,379 images.This dataset gathers Arabic characters, Latin characters, Arabic numerals, Latin numerals, arithmetic operators, set-symbols, comparison symbols, delimiters, etc. Keywords: Image processing, Handwritten mathematical symbols, Documents, Recognition

A method has been proposed to classify handwritten Arabic numerals in its compressed form using partitioning approach, Leader algorithm and Neural network. Handwritten numerals are represented in a matrix form. Compressing the matrix representation by merging adjacent pair of rows using logical OR operation reduces its size in half. Considering each row as a partitioned portion, clusters are formed for same partition of same digit separately. Leaders of clusters of partitions are used to recognize the patterns by Divide and Conquer approach using proposed ensemble neural network. Experimental results show that the proposed method recognize the patterns accurately.

The present study aimed at investigating the underlying cognitive processes and language specificities of three-digit number processing. More specifically, it was intended to clarify whether the single digits of three-digit numbers are processed in parallel and/or sequentially and whether processing strategies are influenced by the inversion of number words with respect to the Arabic digits [e.g., 43: dreiundvierzig ("three and forty")] and/or by differences in reading behavior of the respective first language. Therefore, English- and German-speaking adults had to complete a three-digit number comparison task while their eye-fixation behavior was recorded. Replicating previous results, reliable hundred-decade-compatibility effects (e.g., 742_896: hundred-decade compatible because 7 < 8 and 4 < 9; 362_517: hundred-decade incompatible because 3 < 5 but 6 > 1) for English- as well as hundred-unit-compatibility effects for English- and German-speaking participants were observed, indicating parallel processing strategies. While no indices of partial sequential processing were found for the English-speaking group, about half of the German-speaking participants showed an inverse hundred-decade-compatibility effect accompanied by longer inspection time on the hundred digit indicating additional sequential processes. Thereby, the present data revealed that in transition from two- to higher multi-digit numbers, the homogeneity of underlying processing strategies varies between language groups. The regular German orthography (allowing for letter-by-letter reading) and its associated more sequential reading behavior may have promoted sequential processing strategies in multi-digit number processing. Furthermore, these results indicated that the inversion of number words alone is not sufficient to explain all observed language differences in three-digit number processing.

Numerical representations influence temporal processing. Previous studies have consistently shown that larger numbers are perceived to last longer than smaller ones. However, whether this effect is modulated by the absolute or relative magnitudes of the numbers has yet to be fully understood. Here, participants observed single- and double-digit Arabic numerals in separate experimental blocks and reproduced stimulus duration of 600 or 1200 ms. Our results replicated previous findings that the duration of larger numbers was reproduced longer than that of smaller numbers within each digit set. Although the effect of numerical magnitude across single- and double-digit numerals was found when the numerals were presented for 600 ms, the difference was negligible when they were presented for 1200 ms, suggesting that relative magnitude is an important factor in the number-time interaction in the suprasecond range. These results suggest that contextual influence on number-time interaction may depend on the actual stimulus duration.

Here, new rules of Latin anatomical nomenclature are proposed to deal with cases not covered by existing or other recommended rules. Determiners (e.g., numerals, letters, alphanumeric strings, and Latin names of Greek letters) should follow the noun they specify or limit, just as it is recommended that adjectives should follow the noun they modify. In general, Roman numerals, Latin letters, and Latin names of Greek letters are preferable to Arabic numerals and Greek letters in Latin anatomical terms. It is also noted that the word typus (type) appears to be superfluous and unnecessary in the Latin anatomical nomenclature. Clin. Anat. 30:700-702, 2017. © 2017Wiley Periodicals, Inc.

Handwriting signature is widely used and the main challenge for handwriting recognition is how to obtain comprehensive handwriting information. Triboelectric nanogenerator is sensitive to external triggering force and can be used to record personal handwriting signals and associated characteristics. In this work, micro/nano structure textured TENG acting as a smart self-powered handwriting pad is developed and its effectiveness for handwriting recognition is demonstrated. Three individuals' handwriting signals of English words, Chinese characters and Arabic numerals are acquired by leaf-inspired TENG, and the other three people's handwriting signals of English sentences and the corresponding Chinese sentences are obtained by cylindrical microstructured PDMS based TENG, and these signals exhibit unique features. Combined with the machine learning method, the people's handwriting was successfully identified. The classification accuracies of 99.66%, 93.63%, 91.36%, 99.05%, and 97.73% were reached for English words, Arabic numerals, Chinese characters, English sentences, and the corresponding Chinese sentences, respectively. The results strongly suggested that the textured TENG exhibited great potential in personal handwriting signature identification, security defense, and private information protection applications.

This paper brings into discussion the problem of recognizing Arabic numbers using a monocular camera as the only sensor. When a digital image is presented in this application, optical character recognition (OCR) can be exploited to comprehend numerical data. However, there has been a limited success when applied to the handwritten Arabic (Indian) numbers. This paper aims to overcome this limitation and introduces optical character recognition system based on skeleton matching. The proposed approach is used for handwritten Arabic numbers only. The experimental results indicate the effectiveness of the proposed optical character recognition system even for numbers written in worst case. The right system achieves a recognition rate of 99.3 %.

Automatic license plate (ALP) detection and recognition is an important task for both traffic surveillance and parking management systems, as well as being crucial to maintaining the flow of modern civic life. Various ALP detection and recognition methods have been proposed to date. These methods generally use various image processing and machine learning techniques. In this paper, a cascaded deep learning approach is proposed in order to construct an efficient ALP detection and recognition system for the vehicles of northern Iraq. The license plates in northern Iraq contain three regions, namely a plate number, a city region, and a country region. The proposed method initially employs several preprocessing techniques such as Gaussian filtering and adaptive image contrast enhancement to make the input images more suited to further processing. Then, a deep semantic segmentation network is used in order to determine the three license plate regions of the input image. Segmentation is then carried out via deep encoder-decoder network architecture. The determined license plate regions are fed into two separate convolutional neural network (CNN) models for both Arabic number recognition and the city determination. For Arabic number recognition, an end-to-end CNN model was constructed and trained, whilst for the city recognition, a pretrained CNN model was further fine-tuned. A new license plate dataset was also constructed and used in the experimental works of the study. The performance of the proposed method was evaluated both in terms of detection and recognition. For detection, recall, precision and F-measure scores were used, and for recognition, classification accuracy was used. The obtained results showed the proposed method to be efficient in both license plate detection and recognition. The calculated recall, precision and F-measure scores were 92.10%, 94.43%, and 91.01%, respectively. Moreover, the classification accuracies for Arabic numbers and city labels were shown to be 99.37% and 92.26%, respectively.

Captcha, which can prevent computer programs from attacking websites, has been the most important security technology for many years. The most popularly deployed Captcha is the text-based scheme. The vast majority of the existing text Captchas are designed with English letters and Arabic numerals. Recently, text Captchas with large character sets are being increasingly popular. From the perspective of attackers, larger character set means greater solution space and better theoretical security. However, the security of Captchas with large character sets in real world has never been studied comprehensively. In this article, we introduce a simple, fast, and effective deep learning method to attack these newly emerging Captchas. Taking 11 Chinese Captchas as representatives, we ran our experimental attack on each of them. Our attack achieved high success rates, ranging from 34.7 to 86.9 percent at an average speed of 0.175 seconds on these schemes. All of the results show that the Chinese text Captcha can be easily broken, demonstrating that text Captchas with large character sets are also insecure in existing forms. As a substitute, we proposed a 3D image-based scheme combining semantic comprehension and dragging action. The preliminary experimental results show that it is more robust than current text-based schemes.

[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT AUTHOR'S REQUEST.] Rapidly evolving of Convolutional Neural Networks (CNNs) appeals and endeavors us to explore and discover various CNN model's robustness leading to establishing more effective and efficient models achieved state-of-the-art results on challenging datasets including novel benchmark used in this work. Comparing with most contemporary works, our models outperform the accuracy demonstrated by all existing models. Enhancement performed during this work is not only included model performance but also compromise other vital deep learning aspects such as speed and efficiency. Moreover, a new challenging Arabic handwriting digit dataset is also introduced in this work. We create novel dataset because recently handwritten digit recognition becomes vital scope and it is appealing many researchers. Also the lacking research of using Arabic digits endeavors us to dig deeper by creating our challenge Arabic Handwritten Digits which consists of more than 45,000 samples.

Intelligent inspection in the substation transformer using optical character recognizer has been developing rapidly. Character segmentation from the text line of data plate is an important step for localization and recognition of electrical equipment. However, on-site character segmentation is challenging if the data plate contains multiple languages, especially when the width between Chinese and non-Chinese character differs significantly and the complex environments cause the light reflection and fading. This paper proposes a new method, based on analyzing the connected component and Chinese character's structure, to segment characters from multi-language data plate of substations. The proposed method uses the combination of the HSV color space and multi-scale MSRCP to reduce the effect of illumination and complex background. The proposed method utilized the width of each kind character, the interval between characters and the relationship within the left-right structure Chinese character to improve the segmentation accuracy. Experimental results show that the text lines from the data plate in substation transformer, including Chinese, English, Roman numerals, Arabic numerals and symbols, can be segmented correctly. Results show that the proposed method outperforms two existing character segmentation methods and achieves 99.4% precision in the multi-language data plate dataset.

Handwritten character recognition has been the center of research and a benchmark problem in the sector of pattern recognition and artificial intelligence, and it continues to be a challenging research topic. Due to its enormous application many works have been done in this field focusing on different languages. Arabic, being a diversified language has a huge scope of research with potential challenges. A convolutional neural network (CNN) model for recognizing handwritten numerals in Arabic language is proposed in this paper, where the dataset is subject to various augmentations in order to add robustness needed for deep learning approach. The proposed method is empowered by the presence of dropout regularization to do away with the problem of data overfitting. Moreover, suitable change is introduced in activation function to overcome the problem of vanishing gradient. With these modifications, the proposed system achieves an accuracy of 99.4% which performs better than every previous work on the dataset.

Medical laboratory report is one kind of essential document for health care professionals in patient assessment, diagnosis, and long-term monitoring. Compared with paper files, electronic records are convenient for keeping up to date, complete, and accurate, which is already common in modern medical system. But the recognition from historical medical laboratory reports is still in great needs, especially in developing countries. In this paper, we present a document image processing system used for extracting information from medical laboratory reports. Given an image of medical laboratory report, its table areas and texts are firstly segmented following a top-down pipeline. Then, recognition is undergoing for every text that may contain Arabic numerals, mathematical symbols, and multilingual characters. We evaluate the system on a new dataset of medical laboratory reports that includes scanned images and camera-captured images. Our experiments demonstrate that the proposed system can effectively segment the medical document according to its layout and recognize the texts mixed with multi-type characters and symbols to obtain information from medical laboratory reports. The proposed system and the public dataset will benefit the remote healthcare in developing countries.

ID check is a vital process to verify people identify and to allow entry to different places like universities, check points or a banks. This process is usually done by just looking into the ID or writing it down. To improve the check process and make it quick and easy, our work develops a novel system that extracts the Arabic numbers from the ID picture.The proposed algorithm uses morphological operations more specifically dilation to maximally eliminate non Region Of Interest (ROI) and enhance the (ROI). Moreover, algorithm applies (Speeded Up Robust Features) SUFE algorithm to extract feature points of each image and the correlation based template matching technique to recognise characters.The approach has been evaluated with 17 ID cards pictures taken by amateur people with their mobile and other pictures from the internet. The images were too bright, dark, or from an angle. The algorithm is tested with these pictures and all ID numbers where identified correctly