Number Recognition Research Articles

딥 러닝을 이용한 빌렛 번호 인식 알고리즘

In the steel industry, billet numbers are typically identified before the rolling process because of customers’ different requirements. To identify the billet number, computer vision systems are widely used, because the billet number is marked on the front of the billet by a specialized marking machine at a high temperature. Conventional algorithms, such as rule-based and machine-learning-based algorithms, require features of objects. The features are designed by the user, and they significantly influence the accuracy of the algorithm. To address this problem, deep learning methods have recently been researched. In image processing, the convolutional neural network (CNN) is widely used among the deep learning methods. We propose an end-to-end algorithm using CNN to detect and recognize the billet numbers used in the steel industry. The proposed algorithm consists of four convolutional layers, three pooling layers, two fully connected layers, and a softmax layer. The output of the CNN model consists of the probabilistic values for 18 classes, which include 17 character classes and 1 background class. By using the output of the CNN model, we obtain a character confidence matrix, and by using the score function, the optimal position of the billet number is detected, and the optimal character is classified. Furthermore, we exploit the fact that the billet number consists of four columns and two rows. The experimental results show that the billet number recognition accuracy is maximized as 95.1% in 20 epochs. Using the proposed algorithm will help to increase operation efficiency in the steel industry.

Automatic Recognition of fMRI-Derived Functional Networks Using 3-D Convolutional Neural Networks.

Current functional magnetic resonance imaging (fMRI) data modeling techniques, such as independent component analysis and sparse coding methods, can effectively reconstruct dozens or hundreds of concurrent interacting functional brain networks simultaneously from the whole brain fMRI signals. However, such reconstructed networks have no correspondences across different subjects. Thus, automatic, effective, and accurate classification and recognition of these large numbers of fMRI-derived functional brain networks are very important for subsequent steps of functional brain analysis in cognitive and clinical neuroscience applications. However, this task is still a challenging and open problem due to the tremendous variability of various types of functional brain networks and the presence of various sources of noises. In recognition of the fact that convolutional neural networks (CNN) has superior capability of representing spatial patterns with huge variability and dealing with large noises, in this paper, we design, apply, and evaluate a deep 3-D CNN framework for automatic, effective, and accurate classification and recognition of large number of functional brain networks reconstructed by sparse representation of whole-brain fMRI signals. Our extensive experimental results based on the Human Connectome Project fMRI data showed that the proposed deep 3-D CNN can effectively and robustly perform functional networks classification and recognition tasks, while maintaining a high tolerance for mistakenly labeled training instances. This study provides a new deep learning approach for modeling functional connectomes based on fMRI data.

End-to-end recognition of slab identification numbers using a deep convolutional neural network

This paper proposes a novel algorithm for the end-to-end recognition of slab identification numbers (SINs). In the steel industry, automatic recognition of an individual product information is important for production management. The recognition of SINs in actual factory scenes is a challenging problem due to complicated background and low-quality of characters. Conventional rule-based algorithms were developed to extract information of SINs, but these methods require engineering knowledge and tedious work for parameter tuning. The proposed algorithm employs a data-driven method to overcome these limitations and to handle the challenges for the recognition of SINs. This paper proposes accumulated response map and model-based score function to effectively use the outputs of a deep convolutional neural network. Experiments were thoroughly conducted for industrial data collected from an actual steelworks to verify the effectiveness of the proposed algorithm. Experiment results demonstrate that simultaneous recognition of entire characters in a SIN by optimizing the model-based score function is more effective for the robust performance compared to separated recognition of individual characters.

Research on the number recognition based on athlete number plate image

Research on the number recognition based on athlete number plate image

Recognition Number of The Vehicle Plate Using Otsu Method and K-Nearest Neighbour Classification

The current topic that is interesting as a solution of the impact of public service improvement toward vehicle is License Plate Recognition (LPR), but it still needs to develop the research of LPR method. Some of the previous researchs showed that K-Nearest Neighbour (KNN) succeed in car license plate recognition. The Objectives of this research was to determine the implementation and accuracy of Otsu Method toward license plate recognition. The method of this research was Otsu method to extract the characteristics and image of the plate into binary image and KNN as recognition classification method of each character. The development of the license plate recognition program by using Otsu method and classification of KNN is following the steps of pattern recognition, such as input and sensing, pre-processing, extraction feature Otsu method binary, segmentation, KNN classification method and post-processing by calculating the level of accuracy. The study showed that this program can recognize by 82% from 100 test plate with 93,75% of number recognition accuracy and 91,92% of letter recognition accuracy.

Effect Of A Physical Activity Intervention On Letter And Number Recognition In Preschoolers

There is considerable evidence that physical activity (PA) may improve some components of cognition and academic performance in preschool-aged children. Increasing PA in an effort to enhance learning in preschoolers may be ideal, particularly as this age group experiences a high rate of cognitive development and neurophysiological change. However, the literature on PA and academic performance related outcomes, such as school readiness skills, in preschoolers is limited. PURPOSE: The purpose of this study was to examine the effects of a 12-week PA intervention on school readiness-related cognitive skills (i.e. letter and number recognition) in preschool children. METHODS: Two preschool centers (n=41 children; 53.6% male; mean±sd: age=4.3±0.7; BMI percentile=52.5±26.3) were randomized to a 12-week preschool-based PA intervention (INT; n=19 children) that incorporated short-bout PA lessons embedded into the Massachusetts Early Learning Standards or a health-tracking control group (CON; n=22). School readiness cognitive skills were assessed by recognition of symbols (i.e. letters and numbers) at baseline and at 12-weeks. Chi square and independent t-tests were used to compare baseline characteristics between groups. The effect of the intervention on letter and number recognition was assessed with repeated measures ANOVA. RESULTS: There were no significant baseline differences between groups. Mean change scores for letter recognition (INT: 3.5±6.8, CON: 5.5±4.9) and number recognition (INT: 0.9±4.3, CON 1.7±2.3) were higher in the CON group compared to the INT group. There was not a significant effect of the intervention on pre- to post-scores. CONCLUSION: The 12-week PA intervention utilized in this study did not elicit a significant improvement in the letter and number recognition of preschoolers. Further research may be valuable to examine the benefits of a preschool PA intervention by utilizing longer intervention periods, additional bouts of academically-tailored PA, and more comprehensive measures of preschool cognitive skills.

Soccer player recognition using spatial constellation features and jersey number recognition

Identifying players in soccer videos is a challenging task, especially in overview shots. Face recognition is not feasible due to low resolution, and jersey number recognition suffers from low resolution, motion blur and unsuitable player pose. Therefore, a method to improve visual identification using spatial constellations is proposed here. This method models a spatial constellation as a histogram over relative positions among all players of the team. Using constellation features might increase identification performance but is not expected to work well as a single mean of identification. Thus, this constellation-based recognition is combined with jersey number recognition using convolutional neural networks. Recognizing the numbers on a player’s shirt is the most straight-forward approach, as there is a direct mapping between numbers and players.Using spatial constellation as a feature for identification is based on the assumption that players do not move randomly over the pitch. Players rather follow a tactical role such as central defender, winger, forward, etc. However in soccer, players do not strictly adhere to these roles, variations occur more or less frequently. By learning constellation models for each player, we avoid a categorical assignment of a player to one single tactical role and therefore incorporate each player’s typical behaviour in terms of switching positions.The presented player identification process is expressed as an assignment problem. Here, an optimal assignment of manually labeled trajectories to known player identities is calculated. Using an assignment problem allows for a flexible fusion of constellation features and jersey numbers by combining the respective cost matrices. Evaluation is performed on 14 different shots of six different Bundesliga matches. By combining both modalities, the accuracy is improved from 0.69 to 0.82 when compared with jersey number recognition only.

RESEARCH AND DEVELOPMENT OF NUMBERS RECOGNITION SYSTEM ON THE BASIS OF FUZZY LOGIC IN LabVIEW

A large amount of data are input into the computer. These data can be presented as visual images, sounds (speech), electronic signals, etc. And there is a problem with their recognition. Recognition of graphic symbol, including letters and numbers is the most common one.There are three basic methods of graphic symbols identification: reference, structural and feature ones. For example, programs like FineReader, Readiris, ScanSoft, OmniPage, CuneiForm can be used to recognize the graphic symbols, which are entered into the computer. Significant reliance recognition accuracy of image quality and some distortion of primary image are main disadvantages of these programs and methodsWe can use fuzzy logic or neural network algorithms to improve recognition accuracy.For example neural networks are used in Google technology.We offered the use of numbers recognition system based on fuzzy logic to improve the accuracy and the feature method was proposed.As informative parameters, i.e. recognition features, two positive and two negative transitions were chosen. Transition is defined as the difference between the current local maximum (or minimum) and the previous local minimum (or maximum) number of pixels (intensity) which include images. The number of pixels is counted within vertical sectors into which the number of images is divided and the width of these sectors may vary from 1 to 3 pixels. Another informative parameter is the number of transitions counted in the separation of the image number into the horizontal sectors.Using of Fuzzy Logic Designer in LabVIEW for design and realization of the fuzzy logic system is considered in the article. The recognition system was designed as a software component in package LabVIEW.The research of the developed recognition system in terms of distortion of numbers images by their rotation was held. The angle of this turn may vary from 0 to 10 both clockwise and counterclockwise.By turning clockwise the range of rotation angle is 2-4 times larger than counterclockwise turn. This depends on the width of the sector, which can vary from 1 to 3 pixels. By turning clockwise the range of distortion is narrowed to 1.5 times when the width of sector increases.In the future we plan to optimize the system to increase the allowable angle of image rotation and research the impact of distortion in the form of the geometric size changes.

Maths needs solid foundations

There appears to be a national obsession with numeral recognition and recording of numbers, and there is an alarming rise in inappropriate expectations around what children should be able to achieve.

Gate automation system evaluation

Purpose This study has two purposes. The first is to identify the determinants influencing the selection of a container number recognition system via a quantitative method to thereby establish an evaluation structure. The second purpose is to conduct an empirical study to determine the weights of the criteria and alternatives. Design/methodology/approach The exploratory factor analysis (EFA) and fuzzy analytic hierarchy process (AHP) were applied to determine the evaluation structure and weights of the criteria and alternatives, respectively. Findings An empirical study based on a dedicated terminal at Keelung Port is conducted. The result demonstrates that the radio-frequency identification (RFID) system is a suitable system for the terminal under consideration in this study. Originality/value The value of this study is twofold. First, EFA was applied to extract common factors from a wide questionnaire survey, thereby establishing a hierarchical analysis structure. This method and comprehensive evaluation structure are useful references for both practitioners and researchers to deal with problems of gate automation. Second, fuzzy AHP was used to decide the weights of the hierarchical structure. The weights obtained by this method are more objective and rational as the imprecision expressions in returned samples have been considered and dealt with.

Neural consequences of post-exertion malaise in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome

Post exertion malaise is one of the most debilitating aspects of Myalgic Encephalomyelitis/Chronic Fatigue Syndrome, yet the neurobiological consequences are largely unexplored. The objective of the study was to determine the neural consequences of acute exercise using functional brain imaging. Fifteen female Myalgic Encephalomyelitis/Chronic Fatigue Syndrome patients and 15 healthy female controls completed 30min of submaximal exercise (70% of peak heart rate) on a cycle ergometer. Symptom assessments (e.g. fatigue, pain, mood) and brain imaging data were collected one week prior to and 24h following exercise. Functional brain images were obtained during performance of: 1) a fatiguing cognitive task – the Paced Auditory Serial Addition Task, 2) a non-fatiguing cognitive task – simple number recognition, and 3) a non-fatiguing motor task – finger tapping. Symptom and exercise data were analyzed using independent samples t-tests. Cognitive performance data were analyzed using mixed-model analysis of variance with repeated measures. Brain responses to fatiguing and non-fatiguing tasks were analyzed using linear mixed effects with cluster-wise (101-voxels) alpha of 0.05. Myalgic Encephalomyelitis/Chronic Fatigue Syndrome patients reported large symptom changes compared to controls (effect size ≥0.8, p<0.05). Patients and controls had similar physiological responses to exercise (p>0.05). However, patients exercised at significantly lower Watts and reported greater exertion and leg muscle pain (p<0.05). For cognitive performance, a significant Group by Time interaction (p<0.05), demonstrated pre- to post-exercise improvements for controls and worsening for patients. Brain responses to finger tapping did not differ between groups at either time point. During number recognition, controls exhibited greater brain activity (p<0.05) in the posterior cingulate cortex, but only for the pre-exercise scan. For the Paced Serial Auditory Addition Task, there was a significant Group by Time interaction (p<0.05) with patients exhibiting increased brain activity from pre- to post-exercise compared to controls bilaterally for inferior and superior parietal and cingulate cortices. Changes in brain activity were significantly related to symptoms for patients (p<0.05). Acute exercise exacerbated symptoms, impaired cognitive performance and affected brain function in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome patients. These converging results, linking symptom exacerbation with brain function, provide objective evidence of the detrimental neurophysiological effects of post-exertion malaise.

A Survey on Banknote Recognition Methods by Various Sensors.

Despite a decrease in the use of currency due to the recent growth in the use of electronic financial transactions, real money transactions remain very important in the global market. While performing transactions with real money, touching and counting notes by hand, is still a common practice in daily life, various types of automated machines, such as ATMs and banknote counters, are essential for large-scale and safe transactions. This paper presents studies that have been conducted in four major areas of research (banknote recognition, counterfeit banknote detection, serial number recognition, and fitness classification) in the accurate banknote recognition field by various sensors in such automated machines, and describes the advantages and drawbacks of the methods presented in those studies. While to a limited extent some surveys have been presented in previous studies in the areas of banknote recognition or counterfeit banknote recognition, this paper is the first of its kind to review all four areas. Techniques used in each of the four areas recognize banknote information (denomination, serial number, authenticity, and physical condition) based on image or sensor data, and are actually applied to banknote processing machines across the world. This study also describes the technological challenges faced by such banknote recognition techniques and presents future directions of research to overcome them.

Vision-based hand gesture recognition of alphabets, numbers, arithmetic operators and ASCII characters in order to develop a virtual text-entry interface system

Hand gesture recognition can substitute the use of text-entry interface for human computer interaction. However, it is a challenging task to develop a virtual text-entry interface covering a large number of gesture-based characters. In this paper, 18 new ASCII printable characters have been introduced along with some of the previously introduced characters [A–Z alphabets, 0–9 numbers and four arithmetic operators (add, minus, multiply, divide)]. In addition to some of the efficient existing features, three new features of 15 dimensions have been incorporated to enhance the performance of the system, which are normalized distance between direction extreme, close figure test and direction change ratio. These features are measured for single-stroke as well as multistroke gestures. An experimental analysis has been carried out for selection of optimal features using the statistical analysis techniques such as one-way analysis of variance test, Kruskal–Wallis test, Friedman test in combination with incremental feature selection technique. Furthermore, a comparative study has been carried out for classification of 58 gestures with the new list of features. A comparative analysis has been performed using five classifiers, namely SVM, kNN, Naive Bayes, ANN and ELM. It has been observed that maximum accuracy achieved using the combination of existing and proposed features is 96.95%, as compared to 94.60% accuracy achieved using existing features for classification of 58 gestures.

An effective method for plate number recognition

In this paper, a novel plate number localization algorithm called Secondary Positioning (SP) is proposed and tested on locally acquired data set. The rough position of a plate number is identified at the first stage by searching red light regions in HSV colour space, and then an accurate position of the plate number is localized in the second stage by finding out vertical edge of the plate number. Template matching which is implemented for recognizing individual character incorporates correction coefficient calculated between the templates and testing images. The test dataset includes two subsets from our local plate numbers, namely, 120 images are used for plate number localization, and 80 images for plate number recognition. The precisions obtained from plate number localization and recognition are approximately 75% and 70% respectively, indicating the performances are better than expected compared to the relevant contexts from the literature survey.

Number Recognition in the Saudi License Plates using Classification and Clustering Methods

Number Recognition in the Saudi License Plates using Classification and Clustering Methods

Application of an Intelligent Transportation System in a Travel Time Information System: Safety Assessment and Management

The research presented here evaluated road safety on the road sections included in the Intelligent Traffic Control System of the Podhale Region (ISSRRP) in Poland. This travel time information system consists of a remote traffic microwave sensor, cameras, as well as automatic plate number recognition on national roads with variable message signs and a mobile app to suggest alternative routes in the regional road network. The study analyzed changes in safety caused by transferring traffic volume from national to regional rural and suburban road networks. The assessment of the safety performance was performed with an empirical Bayes study, with periods of three years before and after the implementation of ISSRRP. No changes were identified in the safety performance of the national road network after to the introduction of ISSRRP. However, when the overall network is considered, a potential increase in the number of crashes may be expected, depending on the volume of traffic transferred from national to regional roads, and rural or suburban areas. Therefore, a new approach for system management was proposed, taking into account not only improvement in traffic flow, but also safety performance.

A Novel Electromagnetic Signature Based on RF Identification of Numbers

This paper presents a novel method to identify the numerical digits based on the RF signals. When integrated on dielectric substrate and excited with an electromagnetic wave, every number integrated with metallic strips shows an electromagnetic signature. This signature is unique and characterizes the RCS of the radiating number. Consequently, it could be used for identification purposes. A frequency domain approach was applied to a set of numbers corresponding to the standard numeration. Simulation results obtained with the electromagnetic simulation software CST-MWS® (Computer Simulation Technology-Microwave-studio) are presented and discussed in this paper. The results of this study confirm that the signatures of each number can establish an algorithm for number recognition and identification without errors.

Two-Digit Number Processing during Childhood: A Mental Rotation Task?

The association between numerical and spatial thinking manifests itself throughout various cognitive observations, concerning arithmetic problem solving as well as simple number recognition tasks. The spatial component in numerical thinking is assumed to represent the semantic processing pathway, for example, the mental representation of numerical magnitude. There is an interconnection of semantic, linguistic, and visual/symbolic pathways, which is thought to be stronger in adults than in children. During childhood, the semantic pathway is supposed to play a major part in numerical processing. In this study, we conducted a functional magnetic resonance imaging paradigm combining a mental rotation task with a number processing task. The behavioral data showed the highest error rate for two-digit numbers (not for three-digit numbers). In regard to two-digit numbers, we also found the strongest neuronal activation overlap for number processing and mental rotation in the left intraparietal sulcus. This pattern suggests that the mental inversion of digits in native German speakers might be a spatial function related to mental rotation. In terms of overcoming opposed informational input for symbolic and linguistic representation and therefore impeded informational extraction the spatial processing pathway might play a major role. To conclude, spatial training could be an effective educational approach, supporting the interconnection of numerical processing pathways.

Identification of Football Players based on Generic Fourier Descriptor Applied for the Recognition of Numbers

Abstract There are various possible approaches to the task of automatic identification of football players on video sequences. In the paper the numbers on jerseys are analysed and investigated. Firstly, the algorithm for the localisation of sport shirts and consecutively numbers is described. Secondly, the method for the recognition of the extracted numbers is proposed and experimentally investigated. The approach analysed in the paper applies the Generic Fourier Descriptor for the representation of extracted binary shapes of numbers.

Number Recognition from Captured Images

Number Recognition from Captured Images