Character Recognition Research Articles

Development of an Offline Android-based Test Paper Checker Application for Northwestern Visayan Colleges

This research introduces an offline Android-based Test Paper Checker application designed to automate and streamline the grading process in institutions like Northwestern Visayan Colleges. Using optical character recognition (OCR) and machine learning, the app accurately digitizes and scores handwritten test responses, improving grading speed and accuracy. Key features include student registration, exam creation, paper scanning, correction key generation, and item analysis for question performance. The app functions offline, ensuring accessibility without internet connectivity. By reducing manual errors and processing times, it supports educators in delivering faster feedback and aligns with UNSDGs Goal 4 (Quality Education) and Goal 9 (Innovation).

Transliterating Javanese Script Images to Roman Script using Convolutional Neural Network with Transfer Learning

The Javanese script holds immense cultural significance within Indonesia despite its diminishing usage in contemporary contexts. Its presence remains notable in specific regions of Java and remains integral to many historical documents and texts. Consequently, there is an urgent need for a transliteration system adept at converting Javanese script into contemporary scripts like Roman or Indonesian, thereby contributing to preserving Java's linguistic and cultural legacy. However, reading or transliterating Javanese script can be time-consuming, especially for longer texts, presenting considerable challenges for non-native readers. This study aims to develop an effective transliteration system for converting Javanese script into Roman script. This system addresses the pressing need to preserve Java's linguistic and cultural heritage by facilitating the readability and accessibility of Javanese script, especially for non-native readers. This study introduces an Optical Character Recognition (OCR) system tailored to identify Javanese script characters and transcribe them into Roman characters, explicitly focusing on fundamental nglegena and sandhangan swara characters. Individual characters are isolated by leveraging horizontal and vertical projection techniques, facilitating subsequent classification using a Convolutional Neural Network (CNN) employing transfer learning methodologies. The system's achievement of an impressive average similarity score of 90.78% is noteworthy, with the Xception architecture demonstrating superior efficiency in transliteration tasks. Implementing such a system harbors significant promise in safeguarding the Javanese script and enhancing its accessibility to a broader audience. This research contributes substantially to preserving and propagating Indonesia's rich cultural and linguistic heritage amidst the digital age.

Enhanced Malaysian License Plate Recognition System Using an Improved YOLOv2 Model

License Plate Recognition (LPR) has gained popularity among researchers due to its wide range of applications, including law enforcement, monitoring, and toll gate systems. However, existing LPR systems still require improvements to achieve optimum accuracy and speed. The advancements in Convolutional Neural Network (CNN) variants offer potential solutions for these challenges. This primary aim of this system is to ensure accurate and efficient recognition of the vehicle plate characters using CNN techniques. This research utilizes two CNN network architectures for deep object detection to address the Malaysian License Plate Recognition (MLPR) task. The first network is designed to detect the license plate, while the second is responsible for recognizing the characters on the plate. Both networks are cascaded from the architecture of two-stage YOLOv2, providing promising speed and accuracy. The MLPR system achieved an accuracy of 98.75% and a processing speed of 0.0104 seconds, using a total of 2,200 license plate images. In conclusion, the system adapted from deep object detection techniques presents a promising solution for the MLPR problem, based on the achieved accuracy and speed.

BUILDING DOCUMENT READING ASSISTANT FOR THE VISUALLY IMPAIRED

This study introduces a solution that applies document analysis and recognition technologies to enhance document accessibility for individuals with visual impairments. The objective is to develop an algorithm capable of accurately analyzing the content of document components and converting them into voice format. Leveraging the pre-trained YOLOv8 model for document analysis and optical character recognition technology, the image annotation model uses the AIAnytime API and Pix2Tex technology to extract LaTeX code from images, facilitating the conversion of mathematical formulas into spoken words. The research results demonstrate significant progress in effectively supporting document reading, making a meaningful contribution to the field of assistive technology for the visually impaired.

A Review of Research on Handwritten Chinese Character Recognition with Multi-Feature Fusion

This paper analyzes the progress of handwritten Chinese character recognition technology, from two perspectives: traditional recognition methods and deep learning-based recognition methods. Firstly, the complexity of Chinese character recognition is pointed out, including its numerous categories, complex structure, and the problem of similar characters, especially the variability of handwritten Chinese characters. Subsequently, recognition methods based on feature optimization, model optimization, and fusion techniques are highlighted. The fusion studies between feature optimization and model improvement are further explored, and these studies further enhance the recognition effect through complementary advantages. Finally, the article summarizes the current challenges of Chinese character recognition technology, including accuracy improvement, model complexity, and real-time problems, and looks forward to future research directions.

Handwritten text recognition of historical documents using deep neural network technologies

Handwritten text recognition of historical documents using deep neural network technologies

An Overview of Handwritten Text Recognition Using Generative AI: Cutting-Edge Approaches, Key Challenges, and Future Directions

Recognizing handwritten text involves several stages, including data collection, data preparation, feature extraction, and categorization. It is essential for tasks like document processing, robotic automation, and historical document analysis, given the challenges of variations in size and shape. The survey focuses on enhancing the model's capabilities through data augmentation. Additionally, it delves into the intricate realm of recognizing handwritten numbers, presenting relevant work and emphasizing the use of various methods such as neural networks, k-nearest neighbor (KNN), and Support Vector Machine (SVM) for classifying handwritten numbers. The main goal of this research is to develop a technique using image processing methods to detect handwritten text within an image, addressing research gaps, challenges, and prospects in the field of text recognition.

Enhanced license plate recognition using deep learning and block-based approach

This study investigates the effectiveness of current deep learning techniques in license plate detection and makes essential contributions. Instead of classifying the characters on Turkish license plates with a single classifier, the characters are divided into blocks of numbers and letters using various image processing techniques, and a separate classifier is used for each block. The proposed approach improves character classification accuracy and license plate recognition accuracy. This approach eliminated the possibility of misclassifying similar letters and numbers and improved the character classification accuracy from 95.9% to 99.6%. In addition, a new character feature dataset was created, and a deep learning model was trained on this dataset. Integrating this model into the system increased the classification accuracy to 99.7%. The YOLOv8 object detection model, trained using CUDA technology, achieved a mAP of 98.9%. The overall accuracy of the whole system in license plate recognition reached 97.3%. This study proves the effectiveness of current deep learning methods and the proposed block-based character recognition approach in license plate recognition.

Comparative Approach to De-Noising TEMPEST Video Frames

Analysis of unintended compromising emissions from Video Display Units (VDUs) is an important topic in research communities. This paper examines the feasibility of recovering the information displayed on the monitor from reconstructed video frames. The study holds particular significance for our understanding of security vulnerabilities associated with the electromagnetic radiation of digital displays. Considering the amount of noise that reconstructed TEMPEST video frames have, the work in this paper focuses on two different approaches to de-noising images for efficient optical character recognition. First, an Adaptive Wiener Filter (AWF) with adaptive window size implemented in the spatial domain was tested, and then a Convolutional Neural Network (CNN) with an encoder–decoder structure that follows both classical auto-encoder model architecture and U-Net architecture (auto-encoder with skip connections). These two techniques resulted in an improvement of more than two times on the Structural Similarity Index Metric (SSIM) for AWF and up to four times for the SSIM for the Deep Learning (DL) approach. In addition, to validate the results, the possibility of text recovery from processed noisy frames was studied using a state-of-the-art Tesseract Optical Character Recognition (OCR) engine. The present work aims to bring to attention the security importance of this topic and the non-negligible character of VDU information leakages.

XAI-SALPAD: Explainable deep learning techniques for Saudi Arabia license plate automatic detection

In recent decades, automatic license plate recognition (ALPR) has emerged as a critical application of artificial intelligence in intelligent transportation systems (ITS), addressing a multitude of existing issues. Several algorithms have been developed to improve ALPR accuracy under a variety of conditions, each with its own set of strengths and limitations. The KSA intends to digitize traditional services as part of its Vision 2030 initiative. In this methodology, the ALPR system for accurately recognizing KSA's license plates (LP) leverages deep neural networks (DNNs) as a powerful technique, capable of performing effectively in unconstrained environments with images taken under various conditions, even when the training data is limited. The approach employs YOLOv8 to detect LPs and alphanumeric characters in real-time, followed by using a convolutional neural network (CNN) for character recognition. According to our finding, the YOLOv8 model outperforms other models in the identification of LPs, achieving remarkable accuracy and F1 scores (mAP@0.5 = 0.96 and mAP@0.95 = 0.97). As a result, YOLOv8 was chosen for SLP character recognition. This strategy is appropriate for ITS since it tackles a number of issues, like decreasing car theft and improving public safety. Furthermore, a locally interpretable agnostic model (LIME) offers justifications that help users better understand the model. The system's provision of clear explanations for its decisions not only satisfies performance standards but also corresponds with the increasing need for explainable AI, guaranteeing an efficient and responsible implementation.

Optical music recognition: challenges and future directions

Optical music recognition: challenges and future directions

HubNet: An E2E Model for Wheel Hub Text Detection and Recognition Using Global and Local Features.

Automatic detection and recognition of wheel hub text, which can boost the efficiency and accuracy of product information recording, are undermined by the obscurity and orientation variability of text on wheel hubs. To address these issues, this paper constructs a wheel hub text dataset and proposes a wheel hub text detection and recognition model called HubNet. The dataset captured images on real industrial production line scenes, including 446 images, 934 word instances, and 2947 character instances. HubNet is an end-to-end text detection and recognition model, not only comprising conventional detection and recognition heads but also incorporating a feature cross-fusion module, which improves the accuracy of recognizing wheel hub texts by utilizing both global and local features. Experimental results show that on the wheel hub text dataset, the HubNet achieves an accuracy of 86.5%, a recall of 79.4%, and an F1-score of 0.828, and the feature cross-fusion module increases the accuracy by 2% to 4%. The wheel hub dataset and the HubNet offer a significant reference for automatic detection and recognition of wheel hub text.

A Transparent Pipeline for Identifying Sexism in Social Media: Combining Explainability with Model Prediction

In this study, we present a new approach that combines multiple Bidirectional Encoder Representations from Transformers (BERT) architectures with a Convolutional Neural Network (CNN) framework designed for sexism detection in text at a granular level. Our method relies on the analysis and identification of the most important terms contributing to sexist content using Shapley Additive Explanations (SHAP) values. This approach involves defining a range of Sexism Scores based on both model predictions and explainability, moving beyond binary classification to provide a deeper understanding of the sexism-detection process. Additionally, it enables us to identify specific parts of a sentence and their respective contributions to this range, which can be valuable for decision makers and future research. In conclusion, this study introduces an innovative method for enhancing the clarity of large language models (LLMs), which is particularly relevant in sensitive domains such as sexism detection. The incorporation of explainability into the model represents a significant advancement in this field. The objective of our study is to bridge the gap between advanced technology and human comprehension by providing a framework for creating AI models that are both efficient and transparent. This approach could serve as a pipeline for future studies to incorporate explainability into language models.

Design of automated model for inspecting and evaluating handwritten answer scripts: A pedagogical approach with NLP and deep learning

We address common challenges examiners face, such as accidental question skipping, marking omissions, and potential bias in assessment. These issues often arise due to the necessity of examining scripts in separate sessions, driven by the high volume of examination materials. In response, we propose the implementation of a self-regulating examiner, harnessing contemporary technology to reduce examiner workload and mitigate the possibility of errors. This automated approach aims to ensure fairness and accuracy in evaluating response scripts, offering a promising solution to the challenges encountered by examiners in the field Our study introduces an innovative approach that seamlessly integrates technologies, including Optical Character Recognition (OCR) for text ex- traction, Natural Language Processing (NLP) for keyword analysis, and ma- chine learning for grading. The results of our method are efficiently presented through a user-friendly web application, providing a streamlined and understandable means for examiners to evaluate response scripts.

License Plate Recognition Using Three-Dimensional Rotated Character Recognition and Instance Segmentation by Deep Learning

License plate recognition is currently used in various situations, such as parking lot vehicle management and the tracking of wanted vehicles. These applications require recognition from all possible camera angles. Therefore, we propose a method for license plate recognition in images captured from various camera angles. First, the license plate area is detected from the input image using the YOLOv5 object detection method. The subsequent process can be performed in two manners: by rotating the license plate to the front using projective transformation and trapezoidal correction, or by using graph matching. The first method performs Hough transforms or linear approximations as preprocessing, and then calculates the rectangle surrounding the license plate from the YOLOv5 output. After performing trapezoidal correction, the maximally stable extremal regions (MSER) are used to detect character candidates, and characters are recognized using 3D rotated character recognition. The second method performs character candidate detection and recognition without trapezoidal correction, followed by graph matching. These methods are versatile because they do not depend on the layout of license plates, which varies among countries and regions. Two types of datasets were used: a set of images containing Japanese license plates collected by ourselves, and a set of publicly available images containing Taiwanese license plates. Two recognition rates were evaluated: the license plate character recognition rate and license plate recognition rate in which all characters in one plate are recognized. The license plate character recognition rates using graph matching were 91.9% for the Japanese license plates and 89.3% for the Taiwanese license plates. The license plate recognition rates were 84.0% and 84.5%, respectively.

A BERT-based with fuzzy logic sentimental classifier for sarcasm detection

This study explores the challenging task of sarcasm detection in text, a crucial aspect of sentiment analysis in Natural Language Processing (NLP). Sarcasm, characterized by irony and nuanced language, often complicates the interpretation of emotional tone in text. To address this, the study employs a hybrid model that integrates BERT (Bidirectional Encoder Representations from Transformers) with fuzzy logic. BERT's deep semantic understanding is combined with the sequential processing of LSTM and the flexible decision-making capabilities of fuzzy logic. The model's performance is validated using multiple datasets, demonstrating a significant improvement in accuracy, particularly in detecting subtle and context-dependent sarcasm. This research contributes to the advancement of sentiment analysis, offering a robust framework for handling complex linguistic expressions in various NLP applications.

Automated High-Throughput Atomic Force Microscopy Single-Cell Nanomechanical Assay Enabled by Deep Learning-Based Optical Image Recognition.

Mechanical forces are essential for life activities, and the mechanical phenotypes of single cells are increasingly gaining attention. Atomic force microscopy (AFM) has been a standard method for single-cell nanomechanical assays, but its efficiency is limited due to its reliance on manual operation. Here, we present a study of deep learning image recognition-assisted AFM that enables automated high-throughput single-cell nanomechanical measurements. On the basis of the label-free identification of the cell structures and the AFM probe in optical bright-field images as well as the consequent automated movement of the sample stage and AFM probe, the AFM probe tip could be accurately and sequentially moved onto the specific parts of individual living cells to perform a single-cell indentation assay or single-cell force spectroscopy in a time-efficient manner. The study illustrates a promising method based on deep learning for achieving operator-independent high-throughput AFM single-cell nanomechanics, which will benefit the application of AFM in mechanobiology.

SMS Scam Detection Application Based on Optical Character Recognition for Image Data Using Unsupervised and Deep Semi-Supervised Learning.

The growing problem of unsolicited text messages (smishing) and data irregularities necessitates stronger spam detection solutions. This paper explores the development of a sophisticated model designed to identify smishing messages by understanding the complex relationships among words, images, and context-specific factors, areas that remain underexplored in existing research. To address this, we merge a UCI spam dataset of regular text messages with real-world spam data, leveraging OCR technology for comprehensive analysis. The study employs a combination of traditional machine learning models, including K-means, Non-Negative Matrix Factorization, and Gaussian Mixture Models, along with feature extraction techniques such as TF-IDF and PCA. Additionally, deep learning models like RNN-Flatten, LSTM, and Bi-LSTM are utilized. The selection of these models is driven by their complementary strengths in capturing both the linear and non-linear relationships inherent in smishing messages. Machine learning models are chosen for their efficiency in handling structured text data, while deep learning models are selected for their superior ability to capture sequential dependencies and contextual nuances. The performance of these models is rigorously evaluated using metrics like accuracy, precision, recall, and F1 score, enabling a comparative analysis between the machine learning and deep learning approaches. Notably, the K-means feature extraction with vectorizer achieved 91.01% accuracy, and the KNN-Flatten model reached 94.13% accuracy, emerging as the top performer. The rationale behind highlighting these models is their potential to significantly improve smishing detection rates. For instance, the high accuracy of the KNN-Flatten model suggests its applicability in real-time spam detection systems, but its computational complexity might limit scalability in large-scale deployments. Similarly, while K-means with vectorizer excels in accuracy, it may struggle with the dynamic and evolving nature of smishing attacks, necessitating continual retraining.

Stretchable, conductive Hydrogel-Based triboelectric nanogenerator integrated with deep learning algorithm for character recognition

With the rapid development of flexible electronics, combining wearable devices together with deep learning has been a hot research topic. Conventional TENGs are constrained by their lack of flexibility, hindering their applications on wearable devices. In this work, a highly stretchable and conductive double network PAM/agar/KCl-4 hydrogel was fabricated, with an elastic modulus of 506.67 kPa, and a conductivity of 1.33 S m−1, the hydrogel showed great potential in strain sensing and information transmission. In addition, we combined deep learning algorithm together with the triboelectric nanogenerator which employed the hydrogel to work as its electrode. The deep learning-assisted character recognition accuracy reached 99.11 %. This work is expected to provide new ideas for flexible TENG devices.

Deep Learning Function Implications For Handwritten Character Recognition Using Improved Mean Square Error Function

تستخدم دوال الخسارة لتقييم دقة أداء نموذج التصنيف في تعلم الماكنة. قيمة معدل الخسارة يقيس اذا ما كان قيم التسميات المتوقعة مطابقة الى التسميات الحقيقة او قريبة منها. تعد تمييزالحروف المكتوبة بخط اليد من مشاكل تمييز الانماط. استخدام التعلم العميق سرع التطور الحاصل في تطبيقات تمييز الانماط لتصنيف الصورالمخزونة.تعد الشبكات العصبية التلافيفية ودالة التفعيل الاكثر استخداما في التعلم العميق. اضافة الى استخدام دوال خسارة مختلفة لتقييم أداء النموذج مثل Categorical Cross-Entropy (CCE) و Mean Square Error (MSE). في تقنيات التعلم العميق, حجم قاعدة البيانات يلعب دور مهم للحصول على أداء عالي, ولكن مع MSE التقليدية قيم الخسارة تصل الى الصفر في مراحل مبكرة من عملية التدريب مع قاعدة البيانات الكبيرة الحجم, بالرغم من كون دقة النموذج ما زالت تتحسن. هذه الدراسة تقترح دالة خسارة مطورة عن طريق تحسين MSE. تستند MSE المقترحة على تقسيم مربع الخطأ على مجموع احتمالات التسمية المتوقعة بدل من مجموع عدد العينات. تم استخدام خمسة قواعد بيانات لاختبار اداء النموذج المقترح الشبكات العصبية التلافيفية مع MSE المطوره بالاضافة الى استخدام نموذج VGG16 المعدل. وقواعد البيانات هم AHCD, AIA9K, HIJJA, Self-collected و MNIST. اظهرت نتائج الدراسة تطور واضح في دقة اداء النموذج المقترح مع متوسط مربع الخطأ MSE المطورة مقارنة ب CCE.