Number Recognition Research Articles

Blockchain for tracking serial numbers in money exchanges

SummaryMoney exchange is one of the most common day‐to‐day activities performed by humans in the daily market. This paper presents an approach to money tracking through a blockchain. The proposed approach consists of three main components: serial number localization, serial number recognition, and a blockchain to store all transactions and ownership transfers. The approach was tested with a total of 110 banknotes of different currency types and achieved an average accuracy of 91.17%. We conducted a user study in real‐time with 21 users, and the mean accuracy across all users was 86.42%. Each user gave us feedback on the proposed approach, and most of them welcomed the idea.

숫자 인식 서비스 기반 다층 퍼셉트론에 대한 입력 데이터 축소 알고리즘

숫자 인식 서비스 기반 다층 퍼셉트론에 대한 입력 데이터 축소 알고리즘

Machine Learning-Based Fast Banknote Serial Number Recognition Using Knowledge Distillation and Bayesian Optimization.

We investigated a machine-learning-based fast banknote serial number recognition method. Unlike existing methods, the proposed method not only recognizes multi-digit serial numbers simultaneously but also detects the region of interest for the serial number automatically from the input image. Furthermore, the proposed method uses knowledge distillation to compress a cumbersome deep-learning model into a simple model to achieve faster computation. To automatically decide hyperparameters for knowledge distillation, we applied the Bayesian optimization method. In experiments using Japanese Yen, Korean Won, and Euro banknotes, the proposed method showed significant improvement in computation time while maintaining a performance comparable to a sequential region of interest (ROI) detection and classification method.

A hybrid convolution network for serial number recognition on banknotes

As the sole identity of banknote, serial number has played a crucial role in monitoring the circulation of currencies. Serial number recognition plays an important role in financial market, which requires fast and accurate performances in real applications. In this paper, a hybrid convolution network model has been proposed, in which a dilated-based convolution neural network is employed to improve the recognition accuracy and a quantitative neural network method is developed to speed up the identification process. In dilated-based convolution neural network, the convolution layer and the pooling layer have been replaced by dilated convolution, which can reduce the computation cost. The quantitative neural network based method quantizes the weight parameters to an integer power of two, which transforms the original multiplication operation to a shift operation and can greatly reduce the time. The proposed model was examined and tested on four different banknotes with 35,000 banknote images including RMB, HKD, USD and GBP. The experimental results show that, the proposed model can efficiently improve the recognition accuracy to 99.89% and reduce the recognition time to less than 0.1 ms, and it outperforms the other algorithms on both recognition accuracy and recognition speed.

Pembelajaran Matematika Anak Usia Dini Melalui Media Dadu dan Pengaruhnya ditinjau dari Perkembangan Kognitif

Learning mathematics for early childhood requires ways and media that are suitable for children's development. Mathematics is a scientific discipline that covers almost all levels of education including early childhood education. The characteristic of early childhood learning is learning while playing as embodied in its curriculum at the level of the National Education System. Media dice games can be used in mathematics learning numerical recognition material that is directly related to aspects of children's cognitive development. This study examines the effect of dice media on cognitive development in the mathematics learning of early childhood education number recognition material. This type of research is quantitative causal with statistical tests used including normality test and homogeneity test and hypothesis testing using T test (T-test) or partial test using SPSS version 22 program. The conclusion of this study is that there are influences of media dice game on cognitive aspects learning material for number recognition in learning early childhood mathematics in the 2018/2019 academic year.

The Deficit of Phonological Processing Associated with both Maths and Reading Difficulties Rather than Separate Maths or Reading Difficulties

In this study, we aimed to estimate the effect of phonological processing in the emergence of specific maths or combined maths and reading difficulties during the first year of schooling. We also estimated whether the high level of phonological processing could be a resource for coping with math difficulties. The study was conducted on a large sample of Russian first-graders (N=3296 pupils, mean age 7.3 years, 49% of them were girls). Pupils were tested twice, at the beginning and at the end of the first grade in their level of maths performance, reading performance, phonological processing, and number recognition skills. In each test, four groups of pupils were identified regarding their level of maths and reading performance: a group with mathematical difficulties only (MD), pupils with reading difficulties only (RD), pupils with both maths and reading difficulties (MDRD) and pupils without difficulties (TD). The probability to move into the MD group, the MDRD group and in the TD group was estimated for pupils regarding their group status at Time 1 and their level of phonological processing. Results revealed that at first grade, phonological processing did not correlate with specific maths difficulties, but associated with both maths and reading difficulties. At the same time, a high level of phonological processing may prevent typically developing pupils from moving into the MD group. Moreover, a high level of phonological processing increases the probability to move into the TD group for pupils who had specific maths difficulties at the start of schooling.

The effect of phonological processing on mathematics performance in elementary school varies for boys and girls: Fixed‐effects longitudinal analysis

Investigations into the relations between phonological processing and math performance have yielded contradictory results. These contradictions might be related to small sample sizes and/or cross‐sectional designs of previous studies that limited the possibility of generalizing the findings obtained. The first goal of this study was to estimate the effect of phonological processing on number recognition and math performance during the first year of schooling, controlling for reading achievement. The second goal was to examine whether this effect varied for boys and girls. To achieve these goals, we used a large sample of first‐graders (N = 3296, 49% were girls) in Russia and applied fixed effects longitudinal analysis. Participants were assessed twice, at the beginning and at the end of the first grade (mean age was 7.3 years at Time 1). The item response theory scaling procedure was used to estimate individual scores for math performance, number recognition, phonological processing and reading performance. The results revealed that phonological processing had effects on number recognition and math performance. This effect was more salient for number recognition than for math performance, probably due to the greater involvement of the retrieval strategy in number recognition tasks. The results also demonstrated that the effect of phonological processing on number recognition was stronger for girls than for boys.

Pengembangan Perangkat Pembelajaran di Sentra Persiapan Menggunakan Video Youtube di Pendidikan Anak Usia Dini

Early Childhood Education (PAUD) is one of the platforms that has been provided by the government for toddlers and young childrento be their initial foothold in the world of education. The media is used by the teacher to deliver the subject matter to students so thatlearning activities can take place effectively, efficiently, and attractively. One of the electronic media that can be used in preparationcenters is computer-based learning software. The purpose of this study is to develop a Learning Kit at the Center for Preparing to UseYoutube Videos in Early Childhood Education. This research belongs to the development research. This research method is RecearchAnd Development (R&D). This research method is used to produce certain products and assess the effectiveness of these products.Development of mathematics learning tools in the form of Syllabus, Monthly Activity Plans (RKB), Daily Activity Plans (RKH), andLearning Books preparation material for the second semester of early childhood education aged 4-6 years. The percentage of studentswho completed 19 and 90.4% presented means that students have achieved a predetermined competency that is understanding theproblems related to number recognition. In addition, students also meet the classical completeness criteria, because the percentage ofstudents who complete is 90.4%, so it can be said that overall students have achieved the specified competencies. from these resultsindicate that this learning can be said to be effective because the value of students who complete by 90.4%.

A Multimodal Approach to Enhancing Automobile Security

The continuous increase in use of the automobiles has resulted in increase in vehicular traffic and social vices such as theft, thus necessitating efforts for improving automobile security. As a result of this, various solutions such as satellite monitoring and General Packet Radio Services have been adopted to track vehicle locations over time. This article combines the uniqueness of an automobile plate number and the face(s) of an authorized vehicle driver to enhance the automobile security system, hence eliminating the inefficiency of the existing tag system while reducing the cases of unauthorized automobile use and outright theft. A weighted score approach was used in evaluating the efficiency of the character recognition subsystem with an accuracy of 89.3% recorded. The face recognition subsystem also achieved reliable genuine and an imposter attempt at 0.0 confidence level. The designed system ensured security of an automobile through the combination of a plate number and image recognition, thereby preventing an unauthorized driver from taking a vehicle out of a controlled area.

심층학습 전처리를 통한 도면 문자 인식 성능 개선

An OCR(Optical Character Recognition) based on deep learning for recognition of characters and numbers on drawings for machine maintenance, is proposed. The proposed pre-processing uses mathematical morphology operation to reduce false detection and recognition due to the leading lines and small parts of the drawings. The LSTM model is used to train and infer drawings of parts book. The results of experiments show that the proposed OCR improves the performance by 6.95 % compared to the classical LSTM In addition, the processing speed is improved compared to the conventional approaches.

The repertoire features of T cell receptor β-chain of different age and gender groups in healthy Chinese individuals

T cell immunity is dependent on T cell receptor (TCR) recognition of large numbers of antigenic peptides presented in the context of major histocompatibility complex (MHC) class I and II molecules. To explore whether the TCR β-chain repertoire changes with different ages and genders in healthy Chinese individuals, we analyzed the TCR β-chain repertoire in 154 healthy Chinese individuals by High-throughput sequencing (HTS) with age (age range: 6–70) and each age group contains about 20 individuals (male and female). Here we report that the extent of TCR diversity was not dependent on gender but it was significantly different between age groups. We found that the rearrangement of the V with J genes in T cell receptor β-chain was highest at the age of 21–30. Moreover, we found that the combination of V6-1 and J2-1 gene had the highest expression over a human lifespan. We further identified J2-1 and V7-2 gene expressed higher at the age of 6–10 and 11–20 than other age groups. However, D gene in TCR β-chain was not significantly diverse in different age groups. In addition, the T cell receptor β complementarity-determining region 3 (CDR3) amino acid sequence expressed the highest between the age of 21–30. These results showed different features of TCR β-chain repertoire in different groups of healthy Chinese individuals. In conclusion, it was expected that these TCR repertoires could serve as a useful tool for investigating the role of immune profiling in healthy Chinese individuals. These results stress the importance of considering age as a factor for immune response.

The Impact of an Educational Media Intervention to Support Children’s Early Learning in Rwanda

Children in developing countries often lack sufficient support for early learning skills prior to beginning school. This research evaluates an educational media intervention using an animated cartoon program, Akili and Me. The program was originally created in Tanzania to teach early learning skills. This program was adapted in content and language use in this study in Rwanda. The two-week intervention involved primary school students (mean age = 7.1 years) who were randomized into two groups (intervention and comparison group). The intervention group viewed one Akili and Me episode a day for five days. This viewing was repeated the following week. Similarly, the comparison watched the same amount of television but the content consisted of local popular programs. Baseline and follow-up assessments evaluated 10 areas of early learning, using an adaptation of the International Development and Early Learning Assessment (IDELA), and also children’s media receptivity. At follow-up, children in the intervention program, Akili and Me, had significantly higher scores for counting, number recognition, shape knowledge, letter identification, color identification, body part recognition, health knowledge, and vocabulary. The analyses provide promising evidence that locally produced educational media interventions can impact early learning skills, even among children living in resource-poor communities.

FPGA Implementation of Hand-written Number Recognition Based on CNN

Convolutional Neural Networks (CNNs) are the state-of-the-art in computer vision for different purposes such as image and video classification, recommender systems and natural language processing. The connectivity pattern between CNNs neurons is inspired by the structure of the animal visual cortex. In order to allow the processing, they are realized with multiple parallel 2-dimensional FIR filters that convolve the input signal with the learned feature maps. For this reason, a CNN implementation requires highly parallel computations that cannot be achieved using traditional general-purpose processors, which is why they benefit from a very significant speed-up when mapped and run on Field Programmable Gate Arrays (FPGAs). This is because FPGAs offer the capability to design full customizable hardware architectures, providing high flexibility and the availability of hundreds to thousands of on-chip Digital Signal Processing (DSP) blocks. This paper presents an FPGA implementation of a hand-written number recognition system based on CNN. The system has been characterized in terms of classification accuracy, area, speed, and power consumption. The neural network was implemented on a Xilinx XC7A100T FPGA, and it uses 29.69% of Slice LUTs, 4.42% of slice registers and 52.50% block RAMs. We designed the system using a 9-bit representation that allows for avoiding the use of DSP. For this reason, multipliers are implemented using LUTs. The proposed architecture can be easily scaled on different FPGA devices thank its regularity. CNN can reach a classification accuracy of 90%.

Pengenalan Nomor Ruangan Menggunakan Kamera Berbasis OCR Dan Template Matching

Most important room parameters to easily recognize with provide numbering of room, the number having function as room address. Camera have function to getting room number picture. To recognize of room number used OCR (Optical Characters Recognition) and Template Matching method. The system having purpose to make it easy the room number recognize process using camera. The first process Camera capture result will be identified using OCR system, and cahange image to be ASCII trought several stages, between segmentation, normalisation, feature extraction and recognition process. The image result of OCR, used input as template matching method, this method having several stages between resize, gray, edge canny, histogram, classification, identification image process, and matching with storage image. The result of system tested, getting several parameters should be maximized to increase success rate in room number recognition proccess, between using a standart room number, angle of camera capture is 90o, and luminance 224-230. From setting paramteters, this system can detect and recognize the room number of 93,75% success rate.

Making Of Flash Card And Augmented Reality-Based Application As A Learning Media For Children With Special Needs

This study aims to measure the level of effectiveness of the use of flash cards and applications based on Augmented Reality as a learning aid. The use of Android applications and AR cards that can also function as flash cards has proven to attract interest and increase student motivation in learning. Number recognition material and addition of unit numbers are chosen as the initial experiment stage which aims to introduce the use of assistive devices in the form of tablet PCs or smartphones to the children. Based on the results of the study, both teachers and students felt helped by the existence of this application as a variety of learning media. The research method used in this study is Multiple Baseline Design Across Subjects, while the sampling technique used is census sampling technique.

Vehicle plate number localization using a modified GrabCut algorithm

Vehicle plate number recognition plays an important role in traffic control and surveillance systems. A key stage in any vehicle plate number recognition system is to first locate the vehicle plate number. In this paper, we present a modified GrabCut algorithm for localizing vehicle plate numbers. In contrast with the traditional interactive GrabCut technique, a modified GrabCut algorithm was designed to identify and extract vehicle plate numbers in a completely automatic manner. Our approach extends the use of the traditional GrabCut algorithm with addition of a feature extraction method which uses geometric information to give accurate foreground extraction. Finally, to evaluate the performance of the proposed technique, the localization accuracy is tested with a dataset of 500 vehicle images with vehicle plates from different countries. An accuracy of 99.8% was achieved for the localization of vehicle plates. Comparative analysis is also reported.

End-to-end Billet Identification Number Recognition System

End-to-end Billet Identification Number Recognition System

A curriculum supplement that integrates transmedia to promote early math learning: A randomized controlled trial of a PBS KIDS intervention

This study investigates whether and how the integration of technology and media improves early mathematics outcomes for low-income preschoolers. The study sample was 966 children and 137 teachers in 86 preschool classrooms in California and New York. Preschool classrooms were randomly assigned to one of three conditions: a 10-week PBS KIDS Transmedia Math Supplement condition that supported teachers through an organized sequence of hands-on and digital mathematics activities; a Technology & Media condition that provided teachers with media resources out of context, or a Business-As-Usual (BAU) condition. The media resources were drawn from four PBS KIDS programs: Curious George, The Cat in the Hat Knows a Lot About That!, Sid the Science Kid, and Dinosaur Train. Children in the PBS KIDS Transmedia Math Supplement condition exhibited significantly higher scores on a researcher-developed measure of counting, number recognition and subitizing, shapes, and patterns than did children in the Technology & Media condition (1.43 points, effect size = 0.22, p < 0.001) and the BAU condition (1.51 points, effect size = 0.24, p < 0.001). Marginally significant effects for the same contrasts were found when analyzing the data collected using a standardized assessment of early math knowledge. Teachers in the PBS KIDS Transmedia Math Supplement condition generally implemented the Supplement as intended, and required less coaching support over the course of the study period. In contrast, teachers in the Media & Technology condition continued to use coaches with greater frequency throughout the study period, relying on them particularly for support with the mathematics and with the selection and integration of digital resources to match learning goals. Furthermore, teachers who enacted the PBS KIDS Transmedia Math Supplement significantly improved their beliefs about their own mathematics knowledge and the benefits of technology experiences for preschoolers.

Class Reconstruction in the Space of Natural System Classification

The article deals with the method of natural system classification developed within the paradigm of open systems physics. The method reveals the system ontology of classes by empirical description. Using this method we solved four classification tasks from different knowledge areas: toxicology, semiconductor manufacturing, numbers recognition, biology. For each dataset automatically were obtained: reconstructions of natural system-conditioned features of each class; reconstructions of intensional components; ideal representatives of classes; borders of class areas and referent distribution by class areas.

СТАТИСТИЧНІ КРИТЕРІЇ ОЦІНЮВАННЯ ІНФОРМАТИВНОСТІ ОЗНАК ДЖЕРЕЛ РАДІОВИПРОМІНЮВАННЯ ТЕЛЕКОМУНІКАЦІЙНИХ МЕРЕЖ І СИСТЕМ ПРИ ЇХ РОЗПІЗНАВАННІ

The procedures of comparative analysis using statistical criteria for evaluating the information content of radio sources of telecommunication networks and systems in their classification and recognition as a set of formalized rules for collecting, processing and analyzing the information obtained are considered. In the introduction, the general processes of recognition and classification of sources of radio emission of telecommunication networks are analyzed, the main statistical criteria for evaluating the information content of information features are given. It is noted that most of the mentioned criteria of recognition allow to carry out only ranking of signs and do not provide the solution of the problem of quantitative estimation of their informativeness by the criterion of minimum probability of error or maximum probability of true recognition. With this in mind, a research goal has been formed, which is to develop a procedure for comparative analysis using statistical criteria for evaluating the information content of radio sources of telecommunication networks and systems in their classification and recognition, as a set of formalized rules for collecting, processing and analyzing the information obtained. The study found that the exact value of the probability of error is difficult to obtain, since its estimation requires knowledge of decision thresholds. The integration in the calculation is only possible numerically. Therefore, in order to solve the recognition problem, it is advisable not to use the error probabilities, but their boundaries (upper and lower), which must be strict on the one hand and easily calculated analytically on the other. It should also be borne in mind that the probability of errors and their boundaries are uniquely related to the class distance (classes), which in turn must be clearly related to the probability of true recognition. Based on the analysis of analytical expressions of the statistical criteria for estimating interclass distances, recognition theory establishes mutual analytical relationships between the main criteria of interclass distances. It is substantiated and proposed to solve the problems of recognition by applying the Fali – Semmon transform, where the criterion of optimality is the maximum ratio of the mean differences of the projections of the vectors of the data of the classes to be recognized to the sum of the covariations in the middle of the classes in their projection to the parameter vector, resulting in a modified Fisher ratio. It is also determined that all the criteria considered are designed for a small number of recognition classes, whereas in practice the number of classes and their size is very large and their total number is unknown. Moreover, the recognition process is multi-parameter, which makes it difficult to solve the problems of classification and recognition of objects and sources of radio emission. To overcome this situation, it is proposed to use a criterion based on the coefficient of non-orthogonality of the conditional probability distributions of the probability of a trait, which can be considered as a