Jul 2026· International Scientific Journal of Engineering and Management· 0 citations
TL;DR
A deep learning-based handwritten character recognition system that leverages Convolutional Neural Networks for automatic feature extraction and classification and highlights the effectiveness of deep learning techniques in enhancing recognition performance and reducing classification errors compared to conventional machine learning methods.
Abstract
Handwritten Character Recognition (HCR)
is a significant area of research in the field of pattern
recognition and computer vision, aiming to automatically
identify and convert handwritten text into machine
readable digital formats. Traditional recognition methods
often struggle with variations in writing styles, character
shapes, and image quality. To address these challenges, this
research proposes a deep learning-based handwritten
character recognition system that leverages Convolutional
Neural Networks (CNNs) for automatic feature extraction
and classification. The proposed model is trained on a large
dataset of handwritten characters and incorporates
preprocessing techniques such as image normalization,
noise removal, and segmentation to improve recognition
accuracy. Deep learning enables the system to learn
complex patterns and hierarchical features directly from
raw image data, eliminating the need for manual feature
engineering. Experimental results demonstrate that the
proposed approach achieves high accuracy, robustness, and
generalization across different handwriting styles. The
system effectively recognizes handwritten alphabets,
numerals, and special characters, making it suitable for
applications such as document digitization, postal mail
sorting, bank cheque processing, educational assessment
systems, and historical manuscript preservation. The study
highlights the effectiveness of deep learning techniques in
enhancing recognition performance and reducing
classification errors compared to conventional machine
learning methods. Future enhancements may include
multilingual character recognition, real-time processing
capabilities, and integration with advanced architectures
such as transformers and recurrent neural networks to
further improve accuracy and efficiency.
Index Terms - Handwritten Character Recognition, Deep
Learning, Convolutional Neural Network, Image
Processing, Pattern Recognition, Optical Character
Recognition, Machine Learning.
The Handwritten Digit Recognition System is a machine learning and deep learning–based project developed to accurately identify handwritten numerical digits from input images by using image processing techniques and a Convolutional Neural Network model trained on the MNIST dataset.
R. Rajesh, P. Sravani· International Journal for Re...· 0 citations
The task of Handwritten Text Recognition (HTR) still persists in computer vision and machine learning because there are numerous intra-class and inter-class differences in handwritten characters. In other words, differences in character shapes, stroke widths, styles, slants, spacings, and pen pressures result in great challenges related to automatic recognition. The inability to use HTR technology hinders the access to the information in different forms, e.g., in medical records, legal documents, educational tests, historic archive materials, etc. It is obvious that overcoming this challenge will facilitate more opportunities regarding information availability and automated processing of documents. The goal of this paper is to develop an end-to-end Handwritten Script to Digital Text Predictor (HSDTP) that will be able to convert images of handwritten characters to machine-readable text. In order to implement this goal, the CNN model trained on the MNIST database was developed and optimized for CPU-based hardware implementation without any need in special GPU. Thus, six successive steps should be implemented: image capture, pre-processing, character segmentation, character classification, reconstruction, and deployment of HSDTP as web service. Pre-processing stage involves such operations as grayscale transformation, noise filtering with Gaussian blur, binarization based on the Otsu method and normalization with resizing to 28 × 28 pixels. Character segmentation is done by contour detection; characters are recognized by CNN with prediction scores.
S. Vijayakumar, M. Arun, M.Prakash et al.· 2026 International Conferenc...· 0 citations
A serial cascade of lightweight CNN and spectrum normalized GAN and spectrum normalized GAN, integrating CBAM attention mechanism is proposed, integrating CBAM attention mechanism, with good experimental results.
Offline handwritten word recognition has progressed from whole-word classification to sequence transcription, yet many systems depend on large annotated corpora and exploit lexical regularities over explicit character evidence. This paper presents an alternative formulation for Arabic offline handwritten word recognition, treating characters as spatial objects detected via a Faster Region-Based Convolutional Neural Network rather than symbols generated by a one-dimensional decoder. We construct and release a character-level annotated subset of 2153 handwritten word images from a standard Arabic benchmark, exporting matched detection, sequence, and word-class labels. We also introduce an open-source subword exchange toolkit that creates a controlled structural-generalization benchmark by swapping subwords while preserving handwriting style. Experiments compare the proposed detector against whole-word and sequence-based baselines on both the original held-out split and the perturbed benchmark. Results show sequence models degrade sharply under structural recombination, whereas the proposed detector remains stable, achieving a 26.56% character error rate and 70.0% word accuracy on the perturbed benchmark. These findings demonstrate that explicit character localization provides a robust, data-efficient alternative for Arabic handwritten text recognition in low-resource settings.
Sofiane Medjram, Ruwaidah Saud Alnejaidi· Applied Sciences· 0 citations
The recognition of handwritten Kannada characters remains a challenging research problem due to the script’s complex structural variations, diverse stroke patterns, and high intra-class variability. Conventional character recognition approaches often struggle to achieve reliable performance when confronted with irregular handwriting styles, noisy inputs, and variations in character morphology. This research presents a Deep Neural Framework for Automated Kannada Character Recognition Using Threshold-Driven Feature Analysis, designed to improve script interpretation through integrated preprocessing, adaptive threshold segmentation, feature optimization, and deep neural representation learning. The proposed framework establishes a systematic recognition pipeline where threshold-driven segmentation enhances character boundary identification, while deep neural modeling captures hierarchical spatial and structural features. Existing research on Kannada handwritten recognition has demonstrated the effectiveness of spatial feature extraction, wavelet-based representations, divide-and-conquer strategies, and neural classifiers; however, limitations remain regarding robustness, scalability, and generalized feature learning. The proposed approach addresses these limitations by combining adaptive feature analysis with deep learning principles to create a more efficient recognition architecture. The study theoretically positions deep neural feature learning as a solution for overcoming traditional handcrafted feature constraints and provides insights into automated digitization of Kannada documents, archival systems, and multilingual intelligent applications. The findings indicate that a threshold-driven deep framework can enhance character representation, reduce segmentation errors, and support reliable handwritten script recognition.
Dr. Jone Vakalalabure, Dr. Mereani Naidu· International Journal of Adv...· 0 citations
The automatic partitioning of handwritten signatures into clean and noisy is crucial to reliable authentication in documents. A new approach is presented here that integrates traditional handcrafted texture features such as Histogram of Oriented Gradients (HOG), Local Binary Patterns (LBP) and Gray-Level Co-occurrence Matrix (GLCM) with deep learning models. We have performed experiments on an offline signature dataset DocSign V2 comprising 2,400 grayscale signature images (50 users, equal number of clean and noisy signatures) created for this work. Eight conventional classifiers are tested on individual and combinations of features. The performance of a CNN built for the task and also four transfer learning models - EfficientNetB0, ResNet50, MobileNetV2, and DenseNet121 - are compared. The experiments show the best result of 97.5% was achieved with the HOG-LBP+AdaBoost combination. This is better than the best of deep learning models, MobileNetV2, which reached 95%. This demonstrates that, particularly for small datasets, carefully designed handcrafted features can be used to effectively classify clean and noisy signatures automatically. Moreover, the proposed system can also be used as a post-processing module to enhance signatures for other analysis or verification.
Thejashwini B L, H S Nagendraswamy, Rajashekara M et al.· International journal of com...· 0 citations