Handwritten Recognition Research Articles

Efficient deep learning models for Telugu handwritten text recognition

<p>Optical character recognition (OCR) technology is indispensable for converting and analyzing text from various sources into a format that is editable and searchable. Telugu handwriting presents notable challenges due to the resemblance of characters, the extensive character set, and the need to segment overlapping characters. To segment the overlapping characters, we assess the width of small characters within a word and segment the overlapping characters accordingly. This method is well suited for the segmentation of overlapping compound characters. To address the recognition of similar characters with less training periods we have used ResNet 18 and SqueezeNet models which have achieved character recognition rates of 95% and 94% respectively.</p>

Everyone Leaves a Trace: Exploring Transcriptions of Medieval Manuscripts with Computational Methods

Abstract The topic of this paper is a thirteenth-century manuscript from the French National Library (Paris, BnF français 24428) containing three popular texts: an encyclopedic work, a bestiary and a collection of animal fables. We have automatically transcribed the manuscript using a custom handwritten text recognition (HTR) model for old French. Rather than a content-based analysis of the manuscript’s transcription, we adapt quantitative methods normally used for authorship attribution and clustering to the analysis of scribal contribution in the manuscript. Furthermore, we explore the traces that are left when texts are copied, transcribed and/or edited, and the importance of that trace for computational textual analysis with orthographically unstable historical languages. We argue that the method of transcription is fundamental for being able to think about complex modes of authorship which are so important for understanding medieval textual transmission. The paper is inspired by trends in digital scholarship in the mid-2020s, such as public transcribe-a-thons in the GLAM (Galleries, Libraries, Archives and Museums) sector, the opening up of digitized archival collections with methods such as HTR, and computational textual analysis of the transcriptions.

Balochi Text Segmentation for Establishing Balochi OCR

OCR is considered the fastest way of data entry; the smart conversion of the text data is called handwritten text recognition. Many of the languages possess OCRs and there are still some languages lacking the OCR. Balochi is one of the national languages of the Pakistan country and the most of speakers live in Baluchistan province of Pakistan. Balochi computing is at its infancy and require attention to its many of the approaches to accumulate the level of other languages especially pertaining to the matter of computation. This paper investigates the relation between other Arabic adopting languages and proposes a segmentation algorithm to segment Balochi text paragraphs into lines, lines into words and words into characters. The algorithm has been adopted and fine tuned to produce the accuracy of 95%. The segmentation algorithm will play a role in developing a complex OCR and handwritten recognition of Balochi language.

Performance Comparison of Online Handwritten Telugu Character Recognition Using Various Local and Global Features

As the text is written on a special digitizer or Personal Digital Assistance (PDA) in which a sensor picks up the pen-tip movements along with the pen-up/pen-down switching, its automatic conversion is performed in the online Handwriting Recognition (HR). There are several works related to the online recognition of Devanagari as well as Tamil scripts. Meanwhile, the online recognition works associated with other Indian languages, specifically Telugu, which is complex in its structure together with style, are very few. Our work emphasizes the development of an online handwritten Telugu character recognition system using dominant points with the combination of SVM and performance analysis of various other features. Three classifiers namely, SVM, K-NN and MLP are used to examine the performance of the feature vectors. The proposed research is verified with HP-Lab data available in the UNIPEN format.

Alphabet Handwriting Recognition: From Wood-Framed Hydrogel Arrays Design to Machine Learning Decoding.

Handwriting recognition is a highly integrated system, demanding hardware to collect handwriting signals and software to deal with input data. Nonetheless, the design of such a system from scratch with sustainable materials and an easily accessible computing network presents significant challenges. In pursuit of this goal, a flexible, and electrically conductive wood-derived hydrogel array is developed as a handwriting input panel, enabling recognizing alphabet handwriting assisted by machine learning technique. For this, lignin extraction-refill, polypyrrole coating, and polyacrylic acid filling, endowing flexibility, and electrical conduction to wood are sequentially implemented. Subsequently, these woods are manufactured into a 5×5 array, creating a matrix of signals upon handwriting. Efficient handwritten recognition is then achieved through appropriate manual feature extraction and algorithms with low complexity within a computing network, as demonstrated in this work, the strategic choice of expertise-based feature engineering and simplified algorithms effectively boost the overall model performance on handwriting recognition. With potential adaptability, further applications in customized wearable devices and hands-on healthcare appliances are envisioned.

Single-crystal ferroelectric LiNbO3 thin film-based synaptic devices enabled with tunable domain wall current for neuromorphic computing

Neuromorphic devices can emulate the human brain to process information, which receives lots of attention in the field of artificial intelligence. Synaptic devices based on ferroelectric thin films feature low-power consumption, multifunctionality, and scalability. Among them, ferroelectric charged domain wall (CDW) devices have attracted intensive interest for the implementation of memristive devices due to their ultrahigh integration ability inherited from the nanoscale domain wall thickness. In particular, the preparation of wafer-scale single-crystalline ferroelectric thin films via ion-sliced heterogeneous wafer bonding lays a good foundation for large-scale integration of ferroelectric devices with functional circuits. However, the biomimic synaptic characteristics and the systematic demonstration of synaptic devices are largely unexplored for this material system. Here, we demonstrate a model synaptic device based on a single-crystal ferroelectric LiNbO3 thin film, which provides the desired characteristics for neuromorphic computing. The conductance modulation demonstrates good linearity for efficient neuromorphic computing applications. Simulations using the Modified National Institute of Standards and Technology handwritten recognition dataset prove that LiNbO3-based synaptic devices can operate with an online learning accuracy of 95.1%. The injection and annihilation of the CDW are proposed as the basis of the conductivity modulation by combining with the piezoresponse force microscopy and conductive atomic force microscopy mapping measurements. With the mature fabrication process of the ultrathin high-quality ferroelectric thin films, LiNbO3-based synaptic devices have an extensive application prospect for future neuromorphic computing systems.

Survey Paper on Smart Note Taker

The Smart Note Taker is an innovative tool designed to maximize the entire process of note taking. By doing so, it enables users to make quick and effective notes either handwritten or voice-activated. Using handwriting recognition and voice-to-text technology, the digital tool transcribes handwritten or spoken information as editable digital text. With cloudbased storage and real-time synchronization, users will be able to access and manage their notes anywhere, at any time, and on multiple devices, thus ensuring the smooth integration of the platform into the workflows of the user. In addition, multimedia is supported, and users are enabled to add images, audio, and video to their notes. Moreover, the AI-powered features, such as automatic summarization, content relevance, and suggestions, help users increase productivity. The Smart Note Taker might be taken to be prepared for students, for working professionals and visually impaired users. It should, therefore, be most adapted to the needs of modern note-taking in terms of being convenient, flexible, and accessible

Von OCR und HTR bis NER und LLM

Abstract The presence of the digital is closely linked to the expectation that work processes will be made easier and faster. Considering the barely recognizable increase in speed in editing projects, this article deals with the potentials and challenges of digital tools and processes which, at least in theory, have the potential to speed up editing processes by transferring tasks to the machine that would otherwise be carried out by humans. Using the examples of optical character recognition, handwritten text recognition, named entity recognition and LLM-based dialog systems (AI-based chatbots), the article examines what is associated with the integration of such digital methods and tools into the editing workflow and to what extent the work and transformation processes and, in particular, the roles of human editors and their specific tasks change.

Ancient Yi Script Handwriting Sample Repository

The ancient Yi script has been used for over 8000 years, which can be ranked with Oracle,Sumerian,Egyptian,Mayan and Harappan,and is one of the six ancient scripts in the world. In this article, we collected 2922 handwritten single word samples of commonly used ancient Yi characters. Each character was written by 310 people respectively, with a total of 427,939 valid characters. We completed continuous handwritten text sampling, written by 250 people, with 5 texts per person, covering topics such as Yi astronomy, geography, rituals, and agriculture. In the process of data collection, we proposed an automatic sampling method for ancient Yi script, and completed the automatic cutting and labeling of handwritten samples. Furthermore, we tested the recognition performance of the sorted data set under different deep learning network models. The results show that ancient Yi script has diverse shape structures and rich writing styles, which can be used as a benchmark data set in related fields such as handwritten text recognition and handwritten text generation.

Full Page Handwriting Recognition on CUDA enabled Docker

-Handwritten text recognition is essential for document digitization but often struggles with multiline content. This paper presents an integrated approach using TrOCR, a pre-trained Transformer model, combined with PaddleOCR for enhanced text detection. The integration, optimized with GPU acceleration and multi-threading within a CUDA-enabled Docker environment, addresses the challenges of full-page handwriting recognition. A user-friendly Flask API with a Gradio demo was developed for deployment. Experimental results demonstrate that the system significantly improves the accuracy of multiline text recognition, outperforming existing models. This approach offers a scalable, efficient solution for accurate handwriting recognition in complex document layouts, advancing the field of document digitization.

Fine-grained Automatic Augmentation for handwritten character recognition

With the advancement of deep learning-based character recognition models, the training data size has become a crucial factor in improving the performance of handwritten text recognition. For languages with low-resource handwriting samples, data augmentation methods can effectively scale up the data size and improve the performance of handwriting recognition models. However, existing data augmentation methods for handwritten text face two limitations: (1) Methods based on global spatial transformations typically augment the training data by transforming each word sample as a whole but ignore the potential to generate fine-grained transformation from local word areas, limiting the diversity of the generated samples; (2) It is challenging to adaptively choose a reasonable augmentation parameter when applying these methods to different language datasets. To address these issues, this paper proposes Fine-grained Automatic Augmentation (FgAA) for handwritten character recognition. Specifically, FgAA views each word sample as composed of multiple strokes and achieves data augmentation by performing fine-grained transformations on the strokes. Each word is automatically segmented into various strokes, and each stroke is fitted with a Bézier curve. On such a basis, we define the augmentation policy related to the fine-grained transformation and use Bayesian optimization to select the optimal augmentation policy automatically, thereby achieving the automatic augmentation of handwriting samples. Experiments on seven handwriting datasets of different languages demonstrate that FgAA achieves the best augmentation effect for handwritten character recognition. Our code is available at https://github.com/IMU-MachineLearningSXD/Fine-grained-Automatic-Augmentation

Machine Learning Approach for Arabic Handwritten Recognition

Text recognition is an important area of the pattern recognition field. Natural language processing (NLP) and pattern recognition have been utilized efficiently in script recognition. Much research has been conducted on handwritten script recognition. However, the research on the Arabic language for handwritten text recognition received little attention compared with other languages. Therefore, it is crucial to develop a new model that can recognize Arabic handwritten text. Most of the existing models used to acknowledge Arabic text are based on traditional machine learning techniques. Therefore, we implemented a new model using deep machine learning techniques by integrating two deep neural networks. In the new model, the architecture of the Residual Network (ResNet) model is used to extract features from raw images. Then, the Bidirectional Long Short-Term Memory (BiLSTM) and connectionist temporal classification (CTC) are used for sequence modeling. Our system improved the recognition rate of Arabic handwritten text compared to other models of a similar type with a character error rate of 13.2% and word error rate of 27.31%. In conclusion, the domain of Arabic handwritten recognition is advancing swiftly with the use of sophisticated deep learning methods.

Enhancing the Recognition of Handwritten Arabic Characters through Hybrid Convolutional and Bidirectional Recurrent Neural Network Models

Handwritten Arabic character recognition remains a challenging task in pattern recognition due to the inherent complexities of the cursive script and visual similarities between characters. While deep learning techniques have demonstrated promising results in this domain, further enhancements to the model architecture can drive even greater performance improvements. This study introduces a hybrid deep learning approach that combines Convolutional Neural Networks (CNNs) with Bidirectional Recurrent Neural Networks, specifically Bidirectional Long Short-Term Memory (Bi-LSTM) and Bidirectional Gated Recurrent Units (Bi-GRU). By leveraging the strengths of both convolutional and recurrent neural network components, the proposed models are able to effectively capture spatial features as well as model the temporal dynamics and contextual relationships present in handwritten Arabic text. Experiments conducted on the AHCD and Hijjaa benchmark datasets show that the CNN-Bi-GRU framework achieved state-of-the-art accuracy rates of 97.05% and 91.78% respectively, outperforming previous deep learning-based methods. These results demonstrate the significant performance gains that can be achieved by integrating specialized temporal modeling and contextual representation capabilities into the handwriting recognition pipeline, without the need for explicit segmentation. The findings of this research represent a crucial advancement in the continued development of sophisticated and precise deep learning systems for Arabic handwriting recognition, with broad applications across domains that rely on efficient text extraction from handwritten documents.

Handwritten text recognition of historical documents using deep neural network technologies

Handwritten text recognition of historical documents using deep neural network technologies

Ultrasensitive Flexible Strain Sensor Made with Carboxymethyl-Cellulose-Anchored Carbon Nanotubes/MXene for Machine-Learning-Assisted Handwriting Recognition.

The combination of wearable sensors with machine learning enables intelligent perception in human-machine interaction and healthcare, but achieving high sensitivity and a wide working range in flexible strain sensors for signal acquisition and accurate recognition remains challenging. Herein, we introduced carboxymethyl cellulose (CMC) into a carbon nanotubes (CNTs)/MXene hybrid network, forming tight anchoring among the conductive materials and, thus, bringing enhanced interaction. The silicone-rubber-encapsulated CMC-anchored CNTs/MXene (CCM) strain sensor exhibits an excellent sensitivity (maximum gauge factor up to 71 294), wide working range (200%), ultralow detection limit (0.05%), and outstanding durability (over 10 000 cycles), which is superior to most of the recently reported counterparts also based on a conductive composite film. Moreover, the sensor achieves seamless integration with human skin with the help of a poly(acrylic acid) adhesive layer, successfully obtaining stable and clear waveforms with meaningful profiles from the human body. On this basis, we proposed and realized a novel in-air handwriting recognition method via extracting multiple features of high-quality strain signals assisted by deep neural networks, achieving a high classification accuracy of 98.00 and 94.85% for Arabic numerals and letters, respectively. Our work provides an effective approach for significantly improving strain sensing performance, thereby facilitating innovative applications of flexible sensors.

Handwritten text recognition and information extraction from ancient manuscripts using deep convolutional and recurrent neural network

Handwritten text recognition and information extraction from ancient manuscripts using deep convolutional and recurrent neural network

Enhancing arabic handwriting recognition through optimized deep learning frameworks

Detecting Arabic handwriting is challenging due to letter shapes, intervening segments, and diacritical marks, despite recent advances in pattern recognition. Deep learning architectures ConvNeXt and NFNet-F5 and the meta-heuristic optimization algorithm Aquila Optimizer, inspired by eagle hunting, are used to overcome these challenges. We first review Arabic handwriting recognition literature to determine strengths, weaknesses, and future directions. Next, we describe Arabic handwriting, particularly its interconnectivity, diversity, and many diacritical symbols that make recognition difficult. The Aquila Optimizer is used to optimize CNN hyperparameters in this paper. The goal is to increase the recognition rate and reduce the computational workload. To prove the Aquila Optimizer's efficacy, the ConvNeXt and NFNet-F5 topologies are compared with and without it. The optimized ConvNeXt model has a recognition rate of 98.96%, significantly higher than comparable techniques. It also thoroughly assesses numerous meta-heuristic optimizers and highlights the Aquila Optimizer's potential to improve model performance. This work has enhanced Arabic handwriting recognition by building more precise and efficient models and provides a framework for future research on optimization and applying these models to other scripts like Persian and Urdu.

HTR-VT: Handwritten text recognition with vision transformer

We explore the application of Vision Transformer (ViT) for handwritten text recognition. The limited availability of labeled data in this domain poses challenges for achieving high performance solely relying on ViT. Previous transformer-based models required external data or extensive pre-training on large datasets to excel. To address this limitation, we introduce a data-efficient ViT method that uses only the encoder of the standard transformer. We find that incorporate a Convolutional Neural Network (CNN) for feature extraction instead of the original patch embedding and employ Sharpness-Aware Minimization (SAM) optimizer to ensure that the model can converge towards flatter minima yield notable enhancements. Furthermore, our introduction of the span mask technique, which masks interconnected features in the feature map, acts as an effective regularizer. Empirically, our approach competes favorably with traditional CNN-based models on small datasets like IAM and READ2016. Additionally, it establishes a new benchmark on the LAM dataset, currently the largest dataset with 19,830 training text lines. The code will be publicly available at: https://github.com/YutingLi0606/HTR-VT.

Deep learning-based digitization of Kurdish text handwritten in the e-government system

Many government institutions in developing countries such as the Kurdistan Region of Iraq (KRI) keep a variety of paper-based records that are available in printed or handwritten format. The need for technology that turns handwritten writing into digital text is therefore highly demanded. E-government in developed and developing countries is a crucial facilitator for the provision of such services. This paper aims to develop a deep learning model based on the mask region convolutional neural network (mask-RCNN) to effectively digitize kurdish handwritten text recognition (KHTR). In this research, typical datasets, which includes the isolated handwritten Central Kurdish character images, an extensive database of 40,410 images, and 390 native writers have been produced to determine the developed approach’s performance in terms of identification rates. This approach achieves adequate outcomes in terms of training time and accuracy. The proposed model gives higher performance for detection, localization, and recognition when using a dataset containing many challenges, the results were 80%, 96%, and 87.6 for precision, recall, and F-score respectively. The findings revealed that the proposed model obtained better results compared to other similar works. The accuracy of optical character recognition (OCR) is more than 99%.

Implementing handwritten text recognition using deep learning with TensorFlow: An MNIST dataset approach

Handwritten text recognition (HTR) is a pivotal technology with extensive applications in document digitization, postal automation, and educational tools. This paper delves into the implementing a deep learning-based system for recognizing handwritten digits using TensorFlow and the MNIST dataset. The MNIST dataset, a widely-used benchmark, comprises 60,000 training images and 10,000 testing images of handwritten digits, each standardized to a 28x28 pixel grayscale format. Leveraging the power of Convolutional Neural Networks (CNNs), our model effectively extracts features and classifies digits with high accuracy. The model architecture consists of multiple convolutional layers for feature extraction, pooling layers for dimensionality reduction, and fully connected layers for classification. Preprocessing steps include normalizing pixel values and one-hot encoding the labels, ensuring the data is optimally formatted for training. The TensorFlow framework, known for its robustness and scalability, facilitates the development and deployment of the model. Through a series of experiments, the model demonstrates impressive performance, achieving high accuracy on the MNIST test set. This paper underscores the potential of deep learning in handwritten text recognition. It sets the stage for future enhancements, such as recognizing more complex handwritten texts and optimizing the system for practical applications. The results highlight the effectiveness of deep learning techniques in overcoming the challenges associated with handwritten text recognition, paving the way for advanced, real-world implementations.