Handwritten Alphabets Research Articles

Handwritten Character Recognition: A Comparative Study

Handwriting recognition is a technique used to interpret intelligible handwritten input and convert it into digital text using Machine Learning tools. This research paper provides a comparison of the application of CRNN and CNN for handwriting recognition, using a dataset containing about 370,000 handwritten names. Our experiments demonstrate that the CRNN hybrid model produces the highest accuracy compared to the CNN model. This paper summarises contributions reported on the A-Z Handwritten Alphabets in .csv format dataset for handwritten character recognition. This dataset has been extensively used to validate novel techniques in computer vision. This paper makes a distinction between those works using some kind of data augmentation and works using the original dataset

Handwritten alphabet classification in Tamil language using convolution neural network

Handwritten Alphabet Recognition can be defined as the way of detecting characters from images of Handwritten language alphabets. This is one of the important problems that can be solved by Convolution Neural Networks (CNN). Recent developments in CNN have made it possible to expand this problem area from English character recognition or Numbers recognition to Regional Languages character recognition, there has not been sufficient studies conducted in the domain of regional languages. This study has attempted to give deep learning approach to Tamil Handwritten Alphabets classification. This article aims to develop 3 models of CNN – THAC-CNN1, THAC-CNN2 and THAC-CNN3 to recognize Tamil Handwritten Alphabets and classify them based on its category. Our proposed models use a combination of benchmark dataset and a customized dataset which totals to over 2800 images of different Tamil alphabets after various data augmentation techniques. The proposed models are compared with a popular image classification pre-trained models - VGG-11 and VGG-16. We use the standard classification metric - accuracy to measure the performance of our proposed models. With our dataset and augmentation techniques, one of our models THAC-CNN1 achieves 97% accuracy on the training dataset and 92.5% accuracy on test dataset as opposed to 72% and 73.5% accuracy on training dataset and test dataset by pre-trained models.

RECOGNITION OF REAL-TIME HANDWRITTEN CHARACTERS USING CONVOLUTIONAL NEURAL NETWORK ARCHITECTURE

Pattern recognition, including handwriting recognition, has become increasingly common in everyday life, as is recognizing important files, agreements or contracts that use handwriting. In handwriting recognition, there are two types of methods commonly used, namely online and offline recognition. In online recognition, handwriting patterns are associated with pattern recognition to generate and select distinctive patterns. In handwritten letter patterns, machine learning (deep learning) is used to classify patterns in a data set. One of the popular and accurate deep learning models in image classification is the convolutional neural network (CNN). In this study, CNN will be implemented together with the OpenCV library to detect and recognize handwritten letters in real-time. Data on handwritten alphabet letters were obtained from the handwriting of 20 students with a total of 1,040 images, consisting of 520 uppercase (A-Z) images and 520 lowercase (a-z) images. The data is divided into 90% for training and 10% for testing. Through experimentation, it was found that the best CNN architecture has 5 layers with features (32, 32, 64, 64, 128), uses the Adam optimizer, and conducts training with a batch size of 20 and 100 epochs. The evaluation results show that the training accuracy is between 85, 90% to 89.83% and testing accuracy between 84.00% to 87.00%, with training and testing losses ranging from 0.322 to 0.499. This research produces the best CNN architecture with training and testing accuracy obtained from testing. The developed CNN model can be used as a reference or basis for the development of more complex handwriting pattern recognition models or for pattern recognition in other domains, such as object recognition in computer vision, facial recognition, and other object detection.

Classification of Arabic Alphabets Using a Combination of a Convolutional Neural Network and the Morphological Gradient Method

The field of Optical Character Recognition (OCR) is the process of converting an image of text into a machine-readable text format. The classification of Arabic manuscripts in general is part of this field. In recent years, the processing of Arabian image databases by deep learning architectures has experienced a remarkable development. However, this remains insufficient to satisfy the enormous wealth of Arabic manuscripts. In this research, a deep learning architecture is used to address the issue of classifying Arabic letters written by hand. The method based on a convolutional neural network (CNN) architecture as a self-extractor and classifier. Considering the nature of the dataset images (binary images), the contours of the alphabets are detected using the mathematical algorithm of the morphological gradient. After that, the images are passed to the CNN architecture. The available database of Arabic handwritten alphabets on Kaggle is utilized for examining the model. This database consists of 16,800 images divided into two datasets: 13,440 images for training and 3,360 for validation. As a result, the model gives a remarkable accuracy equal to 99.02%.

Enhancing Handwritten Alphabet Prediction with Real-time IoT Sensor Integration in Machine Learning for Image

Abstract Handwriting Recognition (HWR) is a difficult and varied discipline having applications in a wide range of fields, including banking, education, and administration. This research investigates the two main types of HWR systems: online and offline character recognition. Online HWR entails real-time input utilizing digital pens to capture dynamic handwriting traits. It’s used in contemporary gadgets like tablet computers and for signature verification. Offline HWR, on the other hand, processes scanned documents, making it important in situations such as bank cheque processing and assisting the visually handicapped. The research emphasizes the continuing potential for progress in HWR, notably using machine learning and deep learning approaches. Machine learning, a subset of Artificial Intelligence (AI), is critical in developing character recognition algorithms. The selection of an effective classification model is a vital decision, and the study uses a specific dataset to conduct a comparison analysis of alternative models to help in this process. Such assessments provide useful insights for academics and practitioners, allowing them to make more informed judgements on model development for HWR applications.

Handwritten Letter Recognition using Artificial Intelligence

Abstract: Images are easily processed and analysed by the human brain. When the eye sees a particular image, the brain is able to instantly segment it and recognize its numerous aspects. This project proposes the Deep Learning conceptual models based on Convolutional Neural Network (CNN). A comparison between the algorithms reveals that the handwritten alphabets, classified based on CNNs outperforms other algorithms in terms of accuracy. In this project, different architectures of CNN algorithm are used: Manual Net, Alex Net, LeNet Architecture. These architectures contain a convolution layer, max pooling, flatten, feature selection, Rectifier Linear Unit and fully connected softmax layer respectively. The image dataset with 530 number of training images and 2756 numbers of testing images are used to experiment the proposed network. The best accuracy and loss efficient model will be deployed in the Django framework in order to create a user interface for giving the character to be identified and receiving the output result of identified character. Keywords: Handwritten Recognition, Alphabets, Deep Learning, TensorFlow, CNN, Django.

Writer verification of partially damaged handwritten Arabic documents based on individual character shapes.

Author verification of handwritten text is required in several application domains and has drawn a lot of attention within the research community due to its importance. Though, several approaches have been proposed for the text-independent writer verification of handwritten text, none of these have addressed the problem domain where author verification is sought based on partially-damaged handwritten documents (e.g., during forensic analysis). In this paper, we propose an approach for offline text-independent writer verification of handwritten Arabic text based on individual character shapes (within the Arabic alphabet). The proposed approach enables writer verification for partially damaged documents where certain handwritten characters can still be extracted from the damaged document. We also provide a mechanism to identify which Arabic characters are more effective during the writer verification process. We have collected a new dataset, Arabic Handwritten Alphabet, Words and Paragraphs Per User (AHAWP), for this purpose in a classroom setting with 82 different users. The dataset consists of 53,199 user-written isolated Arabic characters, 8,144 Arabic words, 10,780 characters extracted from these words. Convolutional neural network (CNN) based models are developed for verification of writers based on individual characters with an accuracy of 94% for isolated character shapes and 90% for extracted character shapes. Our proposed approach provided up to 95% writer verification accuracy for partially damaged documents.

Bangla Handwritten Characters Recognition Using Convolutional Neural Network

In the Bangla language, there are 50 complex-shaped characters and working with this huge amount of characters with an appropriate set of features is a tough problem to recognize handwritten characters. Moreover, ambiguity and precision errors are common in handwritten words. Furthermore, among a large number of complex-shaped letters, some are quite similar in shape, making handwritten Bangla characters difficult to recognize. In this work, we proposed a convolutional neural network-based approach for recognizing the handwritten Bangla alphabet. In character recognition, the convolutional neural network (CNN) outperforms most of the other models. However, to guarantee a satisfactory performance, CNNs usually need a great number of samples. Bangla handwriting recognition has been a hot topic for several years, but due to the similarity of many Bangla characters, it's difficult to achieve good results. By training and testing on Bangla character datasets, the model gets a 90.22% validation accuracy for Bangalekha isolated dataset and 93.22% validation accuracy for the Ekush dataset.

One Transistor One Electrolyte-Gated Transistor for Supervised Learning in SNNs

Spiking neural networks (SNNs) are a powerful and efficient information processing approach. However, to deploy SNN on resource-constrained edge systems, compact and low power synapses are required, posing a significant challenge to the conventional silicon-based digital circuits in terms of area- and energy- efficiency. In this study, electrolyte-gated transistors (EGTs) paired with conventional transistors were used as the building blocks to implement SNNs. The one transistor one EGT (1T1E) synapse features heterosynaptic plasticity, which provides a flexible and efficient way to practice supervised learning via spike-timing-dependent plasticity. Based on this method, an SNN with spatiotemporal coding was implemented to recognize the handwritten alphabets, demonstrating 98.3% accuracy at 10% noise level with 5 fJ per synaptic transmission and 1.05 pJ per synaptic programming. These results pave the way for energy-efficiently neuromorphic computing in the future.

Recognition of Urdu Handwritten Alphabet Using Convolutional Neural Network (CNN)

Handwritten character recognition systems are used in every field of life nowadays, including shopping malls, banks, educational institutes, etc. Urdu is the national language of Pakistan, and it is the fourth spoken language in the world. However, it is still challenging to recognize Urdu handwritten characters owing to their cursive nature. Our paper presents a Convolutional Neural Networks (CNN) model to recognize Urdu handwritten alphabet recognition (UHAR) offline and online characters. Our research contributes an Urdu handwritten dataset (aka UHDS) to empower future works in this field. For offline systems, optical readers are used for extracting the alphabets, while diagonal-based extraction methods are implemented in online systems. Moreover, our research tackled the issue concerning the lack of comprehensive and standard Urdu alphabet datasets to empower research activities in the area of Urdu text recognition. To this end, we collected 1000 handwritten samples for each alphabet and a total of 38000 samples from 12 to 25 age groups to train our CNN model using online and offline mediums. Subsequently, we carried out detailed experiments for character recognition, as detailed in the results. The proposed CNN model outperformed as compared to previously published approaches.

Learn2Write: Augmented Reality and Machine Learning-Based Mobile App to Learn Writing

Augmented reality (AR) has been widely used in education, particularly for child education. This paper presents the design and implementation of a novel mobile app, Learn2Write, using machine learning techniques and augmented reality to teach alphabet writing. The app has two main features: (i) guided learning to teach users how to write the alphabet and (ii) on-screen and AR-based handwriting testing using machine learning. A learner needs to write on the mobile screen in on-screen testing, whereas AR-based testing allows one to evaluate writing on paper or a board in a real world environment. We implement a novel approach to use machine learning for AR-based testing to detect an alphabet written on a board or paper. It detects the handwritten alphabet using our developed machine learning model. After that, a 3D model of that alphabet appears on the screen with its pronunciation/sound. The key benefit of our approach is that it allows the learner to use a handwritten alphabet. As we have used marker-less augmented reality, it does not require a static image as a marker. The app was built with ARCore SDK for Unity. We further evaluated and quantified the performance of our app on multiple devices.

Classification and recognition of online hand-written alphabets using Machine Learning Methods

The hand-written alphabet recognition and classification plays an important role in pattern recognition, computer vision as well as image processing. In last few decades, a plethora of applications based on this area are developed such as sign identification, multi lingual learning systems etc. This paper classifies samples of hand-written alphabets into different classes using various machine learning methods. The challenging factor in hand written alphabets recognition lie in variations of style, shape and size of the letters. In this paper a simplified and accurate methodology is proposed based upon engineered features which are evaluated and tested using MatLab tool in comparison to other existing methods. The proposed system achieves a substantial amount of accuracy of 98% as compared to the state of the art approaches.

Object Detection using Convolutional Neural Network and Extended SURF with FIS

The aim of the paper is to detect object using the combination of three algorithms: convolutional neural network (CNN) and extended speeded up robust features (SUFR) and Fuzzy inference system (FIS). Here three types of objects are considered: first, we consider RGB images of hundred different types of objects (for example anchor, laptop airplane, car etc.) taken from benchmark database; second, we take grayscale images of human fingerprint from recognized database; third, Bangla handwritten alphabet from standard database. In this paper we extend the SURF algorithm then the result of the extended SURF is applied in FIS to enhance accuracy of detection. Finally, three algorithms are combined and the accuracy of detection of combined technique is found better than individual one. The combined algorithm provides the average recognition rate for objects of first case as 94.21%, for human finger print as 92.17%%, for Bangla letter as 92.38% and for the Bangla digit as 93.69%.

Tactile sensory coding and learning with bio-inspired optoelectronic spiking afferent nerves

The integration and cooperation of mechanoreceptors, neurons and synapses in somatosensory systems enable humans to efficiently sense and process tactile information. Inspired by biological somatosensory systems, we report an optoelectronic spiking afferent nerve with neural coding, perceptual learning and memorizing capabilities to mimic tactile sensing and processing. Our system senses pressure by MXene-based sensors, converts pressure information to light pulses by coupling light-emitting diodes to analog-to-digital circuits, then integrates light pulses using a synaptic photomemristor. With neural coding, our spiking nerve is capable of not only detecting simultaneous pressure inputs, but also recognizing Morse code, braille, and object movement. Furthermore, with dimensionality-reduced feature extraction and learning, our system can recognize and memorize handwritten alphabets and words, providing a promising approach towards e-skin, neurorobotics and human-machine interaction technologies.

Handwritten Phoenician Character Recognition and its Use to Improve Recognition of Handwritten Alphabets with Lack of Annotated Data

Handwritten Phoenician Character Recognition and its Use to Improve Recognition of Handwritten Alphabets with Lack of Annotated Data

CLASSIFICATION OF PRINTED TEXT AND HANDWRITTEN CHARACTERS WITH NEURAL NETWORKS

Various scanned documents consist of machine printed text as well as handwritten alphabet. The classification of handwritten alphabet and machine printed alphabet in document images is the required process afore character recognition. We can distinguish these two types of alphabet images by their shape structural, statistical and visual difference features. This work proposes the artificial neural network based classification technique for machine printed and handwritten text classification at character level using a set of new features which are combination of statistical, shape structural and visual impression features. The projected technique attained remarkable classification efficiency on two databases; IAM dataset and prepared dataset.

A Survey on Handwritten Character Recognition (HCR) Techniques for English Alphabets

Nowadays Hand written Character Recognition (HCR) is major remarkable and difficult research domain in the area of Image processing. Recognition of Handwritten English alphabets have been broadly studied in the previous years. Presently various recognition methodologies are in well-known utilized for recognition of handwritten English alphabets (character). Application domain of HCR is digital document processing such as mining information from data entry, cheque, applications for loans, credit cards, tax, health insurance forms etc. During this survey we present an outline of current research work conducted for recognition of handwritten English alphabets. In Handwritten manuscript there is no restriction on the writing technique. Handwritten alphabets are complicated to recognize because of miscellaneous human handwriting technique, difference in size and shape of letters, angle. A variety of recognition methodologies for handwritten English alphabets are conferred here alongside with their performance.

Support vector machine for identification of handwritten Gujarati alphabets using hybrid feature space

Gujarati language is used in the western state of Gujarat. Because of its peculiarities, its Optical Character Recognition becomes very difficult. For this language very less work has been done in the area of Optical Character Recognition. In this paper, I have attempted the problem of Optical Character Recognition for handwritten Gujarati alphabets. For this work, forty handwritten alphabets are collected from about one hundred and ninety nine writers. Here; aspect ratio, extent of alphabet, and image subdivision approach has been used as feature space and support vector machine (SVM) has been used for the classification purpose and it gives 86.66 % of performance accuracy. kNN is also used for classification and the result is compared with that of SVM. The paper also describes support vector machine.

Diagonal Based Feature Extraction for Handwritten Alphabets Recognition System Using Neural Network

An off-line handwritten alphabetical character recognition system using multilayer feed forward neural network is described in the paper. A new method, called, diagonal based feature extraction is introduced for extracting the features of the handwritten alphabets. Fifty data sets, each containing 26 alphabets written by various people, are used for training the neural network and 570 different handwritten alphabetical characters are used for testing. The proposed recognition system performs quite well yielding higher levels of recognition accuracy compared to the systems employing the conventional horizontal and vertical methods of feature extraction. This system will be suitable for converting handwritten documents into structural text form and recognizing handwritten names.

Another representation of hand written English alphabets by a sequence of fuzzy sets

In this paper, we describe how to represent lower case hand-written English alphabets by a sequence of two to seven fuzzy sets. Each fuzzy set represents an arc segment of the character and each arc segment is assumed to be a part of an ellipse. The part of an ellipse is defined by five quantities; its short and long radii, its orientation angle, whether it is a part of the lower half or the upper half, and whether it is the full half or a part of a half. Hence, we use the Cartesian product of five fuzzy sets to represent each arc segment. We show that this representation is a translation, rotation, and scaling invariant and that it can be used to generate the hand-written English alphabets. The representation we describe is different from the one proposed earlier by the author and when compared with the previous representation, the one described in this paper simulates more closely the behavior of how one writes English characters.