Handwritten Character Research Articles

Deep convolutional neural network for Lampung character recognition

Recognition of document based, and handwritten characters has recently emerged as highly relevant field of study in the field of digital image processing. The ability to read and write Lampung script is a crucial competency as it helps preserve the language, which is a part of Indonesian culture. This research utilizes data obtained from classified documents and handwritten samples, categorized into eight types. To recognize Lampung characters, deep convolutional neural network (DCNN) architecture is proposed. The novelty of this architecture lies in optimizing document-based and handwritten character recognition to achieve the best performance in terms of accuracy and execution time. The proposed architecture will be compared to principal component analysis (PCA) combined with support vector machine (SVM) to evaluate its results. Experimental results using the DCNN architecture show an average accuracy of 99.3% and an execution time of 283 seconds for all data, while PCA and SVM exhibit an average accuracy of 92.9%. Furthermore, the recognition results for all data from documents and handwritten samples yield satisfactory accuracy of 98.6%. These results make the DCNN architecture suitable for use in recognizing Lampung characters and are expected to make it easier for Lampung people to recognize Lampung character.

Generative innovations for paleography: enhancing character image synthesis through unconditional single image models

Data scarcity in paleographic image datasets poses a significant challenge to researchers and scholars in the field. Unlike modern printed texts, historical manuscripts and documents are often scarce and fragile, making them difficult to digitize and create comprehensive datasets. Recently many innovations have been arrived on single image generative models for natural images but none of them are focused on paleographic character images and other handwritten datasets. In paleographic images like stone inscription characters, maintaining exact shape and structure of character is important unlike natural images. In this paper we propose an unconditional single image generative model, CharGAN for isolated paleographic character images. In the proposed system, augmented images are generated from a single image using generative adversarial networks, while maintaining their structure. Specifically, an external augmentation inducer is used to create higher-level augmentations in the generated images. In addition, the input to the generator is replaced with dynamic sampling from a Gaussian mixture model to make changes to the low-level features. From our experimental results, we infer that these two enhancements make single-image generative models suitable not only for natural images, but also for paleographic character images and other handwritten character datasets, the AHCD dataset, and EMNIST, where the global structure is important. Both the qualitative and quantitative results show that our approach is effective and superior in single-image generative tasks, particularly in isolated character image generation.

Increasing Offline Handwritten Chinese Character Recognition Using Separated Pre-Training Models: A Computer Vision Approach

Offline handwritten Chinese character recognition involves the application of computer vision techniques to recognize individual handwritten Chinese characters. This technology has significantly advanced the research in online handwriting recognition. Despite its widespread application across various fields, offline recognition faces numerous challenges. These challenges include the diversity of glyphs resulting from different writers’ styles and habits, the vast number of Chinese character labels, and the presence of morphological similarities among characters. To address these challenges, an optimization method based on a separated pre-training model was proposed. The method aims to enhance the accuracy and robustness of recognizing similar character images by exploring potential correlations among them. In experiments, the HWDB and Chinese Calligraphy Styles by Calligraphers datasets were employed, utilizing precision, recall, and the Macro-F1 value as evaluation metrics. We employ a convolutional self-encoder model characterized by high recognition accuracy and robust performance. The experimental results demonstrated that the separated pre-training models improved the performance of the convolutional auto-encoder model, particularly in handling error-prone characters, resulting in an approximate 6% increase in precision.

Temporal-spatial cross attention network for recognizing imagined characters

Previous research has primarily employed deep learning models such as Convolutional Neural Networks (CNNs), and Recurrent Neural Networks (RNNs) for decoding imagined character signals. These approaches have treated the temporal and spatial features of the signals in a sequential, parallel, or single-feature manner. However, there has been limited research on the cross-relationships between temporal and spatial features, despite the inherent association between channels and sampling points in Brain-Computer Interface (BCI) signal acquisition, which holds significant information about brain activity. To address the limited research on the relationships between temporal and spatial features, we proposed a Temporal-Spatial Cross-Attention Network model, named TSCA-Net. The TSCA-Net is comprised of four modules: the Temporal Feature (TF), the Spatial Feature (SF), the Temporal-Spatial Cross (TSCross), and the Classifier. The TF combines LSTM and Transformer to extract temporal features from BCI signals, while the SF captures spatial features. The TSCross is introduced to learn the correlations between the temporal and spatial features. The Classifier predicts the label of BCI data based on its characteristics. We validated the TSCA-Net model using publicly available datasets of handwritten characters, which recorded the spiking activity from two micro-electrode arrays (MEAs). The results showed that our proposed TSCA-Net outperformed other comparison models (EEG-Net, EEG-TCNet, S3T, GRU, LSTM, R-Transformer, and ViT) in terms of accuracy, precision, recall, and F1 score, achieving 92.66%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document}, 92.77%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document}, 92.70%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document}, and 92.58%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document}, respectively. The TSCA-Net model demonstrated a 3.65%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document} to 7.49%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document} improvement in accuracy over the comparison models.

The Direction of Slant and Horizontal Stroke as Potential Suggestive Markers for the Determination of Handedness

Abstract Background: The task of document examiners is very challenging, as they need to gather all the relevant facts to make an informed decision regarding the questioned document. In the present study, a total of 300 handwriting samples from 100 different writers (involving 50 right-handed and 50 left-handed) were collected at Bundelkhand University, Jhansi. Slant and stroke, as characteristic features of handwriting, were visually examined under a stereomicroscope to ascertain the handedness. Methods: Slants and Strokes as characteristic features were analyzed in a total of 300 handwriting samples collected from 50 left-handed and 50 right-handed writers of age 17–22 years. A reference text comprising three pangram sentences printed on A4 paper and a blue ball point pen was used for sampling. Results: The data collected during the study were analyzed using a statistical method (Chi-square test) and revealed that the left-handed writers (LHWs) had a significantly (P < 0.05) different writing style than the right-handed ones. Results indicated that LHWs are more likely to make strokes in the right-to-left direction and slant of letters in the backward direction, while right-handed writers make strokes in the left-to-right direction and slant of letters in the forward direction. Conclusions: The findings of the study showed valuable insight and highlighted the significance of the direction of slant and horizontal stroke as potential indicators for the determination of handedness.

Feature Extraction Algorithm based Metaheuristic Optimization for Handwritten Character Recognition

Interest in feature extraction for Handwritten Character Recognition (HCR) has been growing due to numerous algorithms aimed at improving classification accuracy. This study introduces a metaheuristic approach utilizing the Honey Badger Algorithm (HBA) for feature extraction in HCR. The Freeman Chain Code (FCC) is employed for data representation. One challenge with using FCC to represent characters is that extraction results vary depending on the starting points, affecting the chain code's route length. To address this issue, a metaheuristic approach using HBA is proposed to identify the shortest route length and minimize computational time for HCR. The performance metrics of the HB-FCC extraction algorithm are route length and computation time. Experiments on the algorithm use chain code representations from the Center of Excellence for Document Analysis and Recognition (CEDAR) dataset, containing 126 uppercase letter characters. According to the results, the proposed HB-FCC method achieves a route length of 1880.28 and requires only 1.07 seconds to process the entre set of character images.

A Comprehensive Approach for Tamil Handwritten Character Recognition with Feature Selection and Ensemble Learning

A Comprehensive Approach for Tamil Handwritten Character Recognition with Feature Selection and Ensemble Learning

Classifying of Isolated Handwritten Arabic Characters (on-line) Using Machine Learning Methods

The prediction of Arabic handwritten characters is one of the fascinating matters in the area of artificial intelligence (AI) and machine learning, especially in the case of on-line handwriting. In the handwriting recognition system, characters are classified according to their particular categories by utilizing -as input- the values of the features extracted from them. Our major objective in this paper is to classify the isolated characters for online Arabic handwriting (? to ?) using machine learning techniques with the features from EDMs (Edge Direction Matrixes) as opposed to EEDMs (Extended Edge Direction Matrixes). With respect to all classification methods used in this study, EEDMs with machine learning techniques had better performance than EDMs in recognizing isolated Arabic online characters.

Incomplete handwritten Dongba character image recognition by multiscale feature restoration

Incomplete handwritten Dongba character often appears in heritage documents and its recognition is significant for heritage and philology. However, all previous methods always suppose that a complete Dongba character is used as input, and thus fail to achieve satisfactory performance when applied to incomplete Dongba character recognition. In this paper, an end-to-end network (DB2RNet) is proposed for incomplete handwritten Dongba character image recognition by multiscale feature restoration. Specifically, we first develop datasets that contain different levels of incomplete Dongba characters. A restoration module is proposed to restore the input incomplete Dongba character, and then a recognition module is employed to recognize Dongba character. By introducing an inter-module residual connection between the restoration module and recognition module, the DB2RNet can strengthen feature information transmission and boost the recognition performance. In addition, novel multiscale feature blocks are introduced, which can provide more effective texture and contextual information transmission for Dongba character image restoration, and thus yielding better restoration effects and better recognition results. Extensive experiments are conducted on Dongba character, Chinese character and Oracle character datasets and validate the effectiveness, superiority and robustness of our methods. Experiments results demonstrate that our proposed DB2RNet achieves competitive Dongba character restoration and recognition performance and outperforms the current state-of-the-art methods.

Алгоритм стеганографической защиты информации в видеофайлах на основе диффузионно-вероятностной модели с шумоподавлением

The results of a study are presented, the purpose of which was to develop a steganography algorithm for hiding text messages in video files. The algorithm is based on the use of a diffusion-probability model with noise reduction, which is implemented by a deep artificial neural network. The algorithm consists of two parts – for the parties sending and receiving the message. On the transmitting side, the following is carried out: synthesis of handwritten images of symbols (signatures) of the line of the hidden message, alignment of their frequency; applying direct diffusion to signatures, resulting in the generation of a noisy image that is deposited into a video stego container. At the receiving end, signatures are extracted from the video content, back diffusion is performed to obtain signatures of handwritten string characters, which are recognized using a convolutional neural network. The novelty of the research lies in the original developed algorithm for steganographic information protection in video files, as well as in a modified method of signature deposition based on the method of replacing the least significant bits. The method consists of bitwise embedding of bytes characterizing the pixel brightness level in the signature into the same blue brightness digits in a sequence of 8 frames of a video stego container. This method made it possible to significantly reduce the visible changes made to the video content when replacing not the least significant bits, but the middle significant bits in the stego container. This, in turn, provides greater resistance to compression attacks when transmitting information over the stegochannel. The practical significance of the research results lies in the developed software, with the help of which the algorithm for steganographic information protection in video files was tested, which showed high values of the peak signal-to-noise ratio and the index of structural similarity of images when embedding information in the middle bits of the bytes that set the brightness of the pixels of the stego container.

VashaNet: An automated system for recognizing handwritten Bangla basic characters using deep convolutional neural network

Automated character recognition is currently highly popular due to its wide range of applications. Bengali handwritten character recognition (BHCR) is an extremely difficult issue because of the nature of the script. Very few handwritten character recognition (HCR) models are capable of accurately classifying all different sorts of Bangla characters. Recently, image recognition, video analytics, and natural language processing have all found great success using convolutional neural network (CNN) due to its ability to extract and classify features in novel ways. In this paper, we introduce a VashaNet model for recognizing Bangla handwritten basic characters. The suggested VashaNet model employs a 26-layer deep convolutional neural network (DCNN) architecture consisting of nine convolutional layers, six max pooling layers, two dropout layers, five batch normalization layers, one flattening layer, two dense layers, and one output layer. The experiment was performed over 2 datasets consisting of a primary dataset of 5750 images, CMATERdb 3.1.2 for the purpose of training and evaluating the model. The suggested character recognition model worked very well, with test accuracy rates of 94.60% for the primary dataset, 94.43% for CMATERdb 3.1.2 dataset. These remarkable outcomes demonstrate that the proposed VashaNet outperforms other existing methods and offers improved suitability in different character recognition tasks. The proposed approach is a viable candidate for the high efficient practical automatic BHCR system. The proposed approach is a more powerful candidate for the development of an automatic BHCR system for use in practical settings.

Chronological Evolution: Development and Identification of a Marathi Handwritten Character Dataset Using Deep Learning

Chronological Evolution: Development and Identification of a Marathi Handwritten Character Dataset Using Deep Learning

Bangla Handwritten Character Recognition Using Deep Learning Approaches and its Explainability With AI

Bangla Handwritten Character Recognition Using Deep Learning Approaches and its Explainability With AI

On-device Online Learning and Semantic Management of TinyML Systems

Recent advances in Tiny Machine Learning (TinyML) empower low-footprint embedded devices for real-time on-device Machine Learning (ML). While many acknowledge the potential benefits of TinyML, its practical implementation presents unique challenges. This study aims to bridge the gap between prototyping single TinyML models and developing reliable TinyML systems in production: (1) Embedded devices operate in dynamically changing conditions. Existing TinyML solutions primarily focus on inference, with models trained offline on powerful machines and deployed as static objects. However, static models may underperform in the real world due to evolving input data distributions. We propose online learning to enable training on constrained devices, adapting local models toward the latest field conditions. (2) Nevertheless, current on-device learning methods struggle with heterogeneous deployment conditions and the scarcity of labeled data when applied across numerous devices. We introduce federated meta-learning incorporating online learning to enhance model generalization, facilitating rapid learning. This approach ensures optimal performance among distributed devices by knowledge sharing. (3) Moreover, TinyML’s pivotal advantage is widespread adoption. Embedded devices and TinyML models prioritize extreme efficiency, leading to diverse characteristics ranging from memory and sensors to model architectures. Given their diversity and non-standardized representations, managing these resources becomes challenging as TinyML systems scale up. We present semantic management for the joint management of models and devices at scale. We demonstrate our methods through a basic regression example and then assess them in three real-world TinyML applications: handwritten character image classification, keyword audio classification, and smart building presence detection. The results confirm the effectiveness of our approaches from various perspectives, such as accuracy improvement, resource savings, and engineering effort reduction.

Chronological Evolution: Development and Identification of an Odia Handwritten Character Dataset Using Deep Learning

Chronological Evolution: Development and Identification of an Odia Handwritten Character Dataset Using Deep Learning

Off-line identifying Script Writers by Swin Transformers and ResNeSt-50

In this work, we present two advanced models for identifying script writers, leveraging the power of deep learning. The proposed systems utilize the new vision Swin Transformer and ResNeSt-50. Swin Transformer is known for its robustness to variations and ability to model long-range dependencies, which helps capture context and make robust predictions. Through extensive training on large datasets of handwritten text samples, the Swin Transformer operates on sequences of image patches and learns to establish a robust representation of each writer’s unique style. On the other hand, ResNeSt-50 (Residual Neural Network with Squeeze-and-Excitation (SE) and Next Stage modules), with its multiple layers, helps in learning complex representations of a writer’s unique style and distinguishing between different writing styles with high precision. The SE module within ResNeSt helps the model focus on distinctive handwriting characteristics and reduce noise. The experimental results demonstrate exceptional performance, achieving an accuracy of 98.50% (at patch level) by the Swin Transformer on the CVL database, which consists of images withcursively handwritten German and English texts, and an accuracy of 96.61% (at page level) by ResNeSt-50 on the same database. This research advances writer identification by showcasing the effectiveness of the Swin Transformer and ResNeSt-50. The achieved accuracy underscores the potential of these models to process and understand complex handwriting effectively.

EVALUATION OF GEOMETRIC MORPHOMETRIC APPROACH FOR ETHNICITIES DISCRIMINATION USING HANDWRITTEN NUMERAL CHARACTERS

Handwriting evidence is a valuable source for authorship identification, an important aspect in investigating crimes such as murder, suicide, illegal drug trafficking, kidnapping, and document forgery. It relies heavily on the examination of written characters that make the document. However, specific studies on the handwritten numeral characters are scarce despite being crucial in assisting investigators in solving crimes. Hence, this study is aimed to gauge the possibility to discriminate authors according to their ethnicities by means of their handwritten numeral characters using a novel Geometric Morphometric (GMM) technique. Handwritten numeral characters collected from 30 individuals from three main different ethnic groups in Malaysia; Malay, Chinese and Indian were first digitised and landmarked using GMM software. Cluster patterns can be observed in the Principal Component Analysis (PCA) score plots, belonging exclusively to the three different ethnic groups. Significant differences (p<0.0001) were discovered in handwritten numerals characters 3, 4, 5, 7 and 9 amongst the three ethnicities when tested using Procrustes ANOVA, which signifying that it is possible to discriminate authors according to their ethnicities using their handwritten numeral characters. However, more sophisticated meta-analyses are needed in order to find the most effective technique for determining and discriminating the author's ethnicity.

ML & DL approach for Handwritten Character Recognition for Gujarati Characters

ML & DL approach for Handwritten Character Recognition for Gujarati Characters

Hindi-Handwritten-Character-Recognition using Deep learning

Abstract: Recognizing handwritten Hindi characters poses a significant challenge in the realms of machine learning and computer vision, particularly in the context of India's accelerating digitization. To address this, accurate and efficient algorithms are imperative for applications ranging from document analysis to postal automation and data entry. Leveraging the advancements in deep learning, we propose a novel approach to Hindi Handwritten Character Recognition. Our method employs a combination of Convolutional Neural Networks (CNNs) to extract image features and Recurrent Neural Networks (RNNs) to capture temporal dependencies within character sequences. Through rigorous evaluation on a standard benchmark dataset, our approach achieves state-of-the-art recognition accuracy. Furthermore, we validate its practical utility by successfully recognizing handwritten postal addresses on envelopes and other real-world applications. This research offers a promising solution to the challenges of Hindi Handwritten Character Recognition, with potential implications for advancing the digitization efforts not only in India but also in analogous regions.

Gujarati handwritten character and modifiers recognition using deep hybrid classifier

Gujarati handwritten character and modifiers recognition using deep hybrid classifier