COMPARATIVE STUDY OF FONT RECOGNITION USING CONVOLUTIONAL NEURAL NETWORKS AND TWO FEATURE EXTRACTION METHODS WITH SUPPORT VECTOR MACHINE

As the prevalence of online services and platforms continues to grow, so too does the threat posed by automated attacks seeking to circumvent security measures such as CAPTCHA (Completely Automated Public Turing test to tell Computers and Humans Apart) mechanisms. This research paper delves into the various attack vectors and vulnerabilities inherent in CAPTCHA systems, aiming to provide a comprehensive understanding of the methodologies employed by attackers to bypass these safeguards. Through an investigation into machine learning-based algorithms, optical character recognition (OCR) techniques, and adversarial attacks, this study sheds light on the evolving landscape of CAPTCHA circumvention strategies. By analyzing the strengths and limitations of each attack vector, as well as the potential countermeasures available, this research contributes to the ongoing efforts to enhance the resilience of CAPTCHA systems against automated threats. Ultimately, the findings of this study serve to inform the development of more robust and adaptive CAPTCHA designs, thereby bolstering cyber security in the digital realm. This research paper focuses on investigating attack vectors and vulnerabilities in CAPTCHA mechanisms, including machine learning-based algorithms, OCR techniques, and adversarial attacks. It emphasizes the importance of understanding these methodologies to inform the development of more resilient CAPTCHA designs that can secure online platforms from various attacks.

Although font recognition is a fundamental issue in the field of document analysis and recognition, it was usually ignored in the past. With the development of optical character recognition (OCR), font recognition becomes more and more important. This paper proposed a well-designed convolutional neural network (CNN) architecture for traditional Mongolian font recognition by means of a single word. To be specific, the whole word image is regarded as input of CNN. Hence, the word images should be normalized into the same size before being inputted into CNN. By comparison, an appropriate aspect ratio for the traditional Mongolian word images has been determined. Experimental results demonstrate that the proposed CNN architecture outperforms three classic CNN architectures, including LeNet-5, AlexNet and GoogLeNet. Therefore, the proposed CNN is much more suitable for the task of the traditional Mongolian font recognition in the way of a single word.

This paper presents a handwritten document recognition system based on the convolutional neural network technique. In today’s world, handwritten document recognition is rapidly attaining the attention of researchers due to its promising behavior as assisting technology for visually impaired users. This technology is also helpful for the automatic data entry system. In the proposed system prepared a dataset of English language handwritten character images. The proposed system has been trained for the large set of sample data and tested on the sample images of user-defined handwritten documents. In this research, multiple experiments get very worthy recognition results. The proposed system will first perform image pre-processing stages to prepare data for training using a convolutional neural network. After this processing, the input document is segmented using line, word and character segmentation. The proposed system get the accuracy during the character segmentation up to 86%. Then these segmented characters are sent to a convolutional neural network for their recognition. The recognition and segmentation technique proposed in this paper is providing the most acceptable accurate results on a given dataset. The proposed work approaches to the accuracy of the result during convolutional neural network training up to 93%, and for validation that accuracy slightly decreases with 90.42%.

Scene texts refer to arbitrary text found in images captured by cameras in real-world settings. The tasks of text detection and recognition are critical components of computer vision, with applications spanning scene understanding, information retrieval, robotics, and autonomous driving. Despite significant advancements in deep learning methods, achieving accurate text detection and recognition in complex images remains a formidable challenge for robust real-world applications. Several factors contribute to these challenges. First, the diversity of text shapes, fonts, colors, and styles complicates detection efforts. Second, the myriad combinations of characters, often with unstable attributes, make complete detection difficult, especially when background interruptions obscure character strokes and shapes. Finally, effective coordination of multiple sub-tasks in end-to-end learning is essential for success. This research aimed to tackle these challenges by enhancing text discriminative representation. This study focused on two interconnected problems: Scene Text Recognition (STR), which involves recognizing text from scene images, and Scene Text Detection (STD), which entails simultaneously detecting and recognizing multiple texts within those images. This research focuses on implementing and evaluating the Efficient and Accurate Scene Text Detector (EAST) algorithm for text detection and recognition in natural scene images. The study aims to compare the performance of three prominent Optical Character Recognition (OCR) techniques—TesseractOCR, PaddleOCR, and EasyOCR. The EAST model was applied to a series of sample test images, and the results were visually represented with bounding boxes highlighting the detected text regions. The inference times for each image were recorded, highlighting the algorithm's efficiency, with average times of 0.446, 0.439, and 0.440 seconds for the respective test images. These results indicate that the EAST algorithm is accurate and operates in real-time, making it suitable for applications requiring immediate text recognition.

Character recognition and classification is considered one of the most important parts of current LPR systems. Because of low recognition quality and poor robustness of traditional character recognition techniques, those techniques were gradually replaced by powerful deep learning modules such as convolutional neural networks. Convolutional neural networks (CNNs) show satisfying ability in character recognition and outperform most of other available models. Since Chinese license plates contain Chinese characters in addition to ordinary alphanumeric characters, a robust, powerful, and efficient CNN is needed to accomplish character recognition task efficiently. In this paper, we have proposed an efficient CNN model based on Darknet architecture to perform character recognition. Through convolutional and max pooling layers, features of input character images will be extracted and then sent to softmax layer for classification. To avoid overfitting problem in the training process, the dropout regularization technique is adopted. We have used a dataset of 84,000 character images for training and testing our model. The experimental result shows satisfactory outputs and eventually achieves test accuracy of 99.69%.

Map labels provide valuable geographic information by annotating geographic phenomenona with text descriptions. However, many interesting and useful maps are only available as images and hence this information is not readily accessible in a Geographic Information System (GIS). Previous work on text recognition in maps considers maps as a special type of image to be processed using Optical Character Recognition (OCR) techniques and does not pay attention to the typical workflows in a GIS. As a result, to convert map labels into machine-readable text, a user has to switch between OCR and GIS software, transform the detected text locations from the image coordinates (in OCR) to the map coordinates (in GIS), and apply data import/export procedures. This tedious process limits the opportunity to access text information in maps. This paper presents ArcStrabo, an integration of our previous text recognition work and a GIS, which uses a GIS user interface, workflows, and data types to enable efficient training of text recognition algorithms for converting map labels to a table of geographic names. We show that ArcStrabo facilitates map digitization processes, eliminates the need for GIS users to learn additional OCR tools, and does not require manual data export/import procedures between GIS and OCR software.

Text recognition in natural scene images is a challenging problem in computer vision. Different than the optical character recognition (OCR), text recognition in natural scene images is more complex due to variations in text size, colors, fonts, orientations, complex backgrounds, occlusion, illuminations and uneven lighting conditions. In this paper, we propose a segmentation-free method based on a deep convolutional recurrent neural network to solve the problem of cursive text recognition, particularly focusing on Urdu text in natural scenes. Compared to the non-cursive scripts, Urdu text recognition is more complex due to variations in the writing styles, several shapes of the same character, connected text, ligature overlapping, stretched, diagonal and condensed text. The proposed model gets a whole word image as an input without pre-segmenting into individual characters, and then transforms into the sequence of the relevant features. Our model is based on three components: a deep convolutional neural network (CNN) with shortcut connections to extract and encode the features, a recurrent neural network (RNN) to decode the convolutional features, and a connectionist temporal classification (CTC) to map the predicted sequences into the target labels. To increase the text recognition accuracy further, we explore deeper CNN architectures like VGG-16, VGG-19, ResNet-18 and ResNet-34 to extract more appropriate Urdu text features, and compare the recognition results. To conduct the experiments, a new large-scale benchmark dataset of cropped Urdu word images in natural scenes is developed. The experimental results show that the proposed deep CRNN network with shortcut connections outperform than other network architectures. The dataset is publicly available and can be downloaded from <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://data.mendeley.com/datasets/k5fz57zd9z/1</uri> .

Functional imaging has successfully been applied to capture functional changes in the pathological tissues of a body in recent years. Nuclear medicine functional imaging has been used to acquire information about areas of concerns (e.g. lesions and organs) in a non‐invasive manner, enabling semi‐automated or automated decision‐making for disease diagnosis, treatment, evaluation, and prediction. Focusing on functional nuclear medicine images, in this study, the authors review existing work on the classification of single‐photon emission computed tomography, positron emission tomography, and their hybrid modalities with computed tomography and magnetic resonance imaging images by using convolutional neural network (CNN) techniques. Specifically, they first present an overview of nuclear imaging and the CNN technique, such as nuclear imaging modalities, nuclear image data format, CNN architecture, and the main CNN classification models. According to the diseases of concern, they then classify the existing CNN‐based work on the classification of functional nuclear images into three different categories. For the typical work in each of these categories, they present details about their research objectives, adopted CNN models, and achieved main results. Finally, they discuss research challenges and directions for developing technological solutions to classify nuclear medicine images based on the CNN technique.

Most Latin word and character recognition systems have numerous classification algorithms to automate the identification of handwritten characters. However, despite extensive research on various Latin-based languages, relatively limited attention has been given to the recognition of handwritten text in Nigerian indigenous languages. Handwritten character recognition is challenging due to the variations in shape and style. In addition to that, these characters naturally vary among individuals. This work presents an offline character recognition system for Igbo handwritten alphabets using convolutional neural networks. In this study, a dataset comprising 9,000 character images was created, with 250 distinct samples for each of the 36 letters in the Igbo alphabet. Of these, 200 samples per character (a total of 7,200 images) were pre-processed, binarized, and normalized for training two Convolutional Neural Network (CNN) models. The remaining 50 samples per character (1,800 images) were reserved for testing. Model 1 and Model 2 achieved recognition accuracies of 83.5% and 85.0%, respectively. The proposed approach demonstrates potential for enhancing the automation and digitization of handwritten Igbo texts into machine-readable formats. Keywords: Optical Character Recognition, Igbo Handwritten Characters, Nigerian, Indigenous, Convolutional Neural Network Journal Reference Format: Woods, N.C. &amp; Oyedeji, O.A. (2025): Offline Handwritten Character Recognition for Igbo Alphabets Using Convolutional Neural Networks. Journal of Behavioural Informatics, Digital Humanities and Development Rese Vol. 11 No. 2. Pp 115-126 https://www.isteams.net/behavioralinformaticsjournal dx.doi.org/10.22624/AIMS/BHI/V11N2P11

Millions of clinicians rely daily on automated preliminary electrocardiogram (ECG) interpretation. Critical comparisons of machine learning-based automated analysis against clinically accepted standards of care are lacking. To use readily available 12-lead ECG data to train and apply an explainability technique to a convolutional neural network (CNN) that achieves high performance against clinical standards of care. This cross-sectional study was conducted using data from January 1, 2003, to December 31, 2018. Data were obtained in a commonly available 12-lead ECG format from a single-center tertiary care institution. All patients aged 18 years or older who received ECGs at the University of California, San Francisco, were included, yielding a total of 365 009 patients. Data were analyzed from January 1, 2019, to March 2, 2021. A CNN was trained to predict the presence of 38 diagnostic classes in 5 categories from 12-lead ECG data. A CNN explainability technique called LIME (Linear Interpretable Model-Agnostic Explanations) was used to visualize ECG segments contributing to CNN diagnoses. Area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were calculated for the CNN in the holdout test data set against cardiologist clinical diagnoses. For a second validation, 3 electrophysiologists provided consensus committee diagnoses against which the CNN, cardiologist clinical diagnosis, and MUSE (GE Healthcare) automated analysis performance was compared using the F1 score; AUC, sensitivity, and specificity were also calculated for the CNN against the consensus committee. A total of 992 748 ECGs from 365 009 adult patients (mean [SD] age, 56.2 [17.6] years; 183 600 women [50.3%]; and 175 277 White patients [48.0%]) were included in the analysis. In 91 440 test data set ECGs, the CNN demonstrated an AUC of at least 0.960 for 32 of 38 classes (84.2%). Against the consensus committee diagnoses, the CNN had higher frequency-weighted mean F1 scores than both cardiologists and MUSE in all 5 categories (CNN frequency-weighted F1 score for rhythm, 0.812; conduction, 0.729; chamber diagnosis, 0.598; infarct, 0.674; and other diagnosis, 0.875). For 32 of 38 classes (84.2%), the CNN had AUCs of at least 0.910 and demonstrated comparable F1 scores and higher sensitivity than cardiologists, except for atrial fibrillation (CNN F1 score, 0.847 vs cardiologist F1 score, 0.881), junctional rhythm (0.526 vs 0.727), premature ventricular complex (0.786 vs 0.800), and Wolff-Parkinson-White (0.800 vs 0.842). Compared with MUSE, the CNN had higher F1 scores for all classes except supraventricular tachycardia (CNN F1 score, 0.696 vs MUSE F1 score, 0.714). The LIME technique highlighted physiologically relevant ECG segments. The results of this cross-sectional study suggest that readily available ECG data can be used to train a CNN algorithm to achieve comparable performance to clinical cardiologists and exceed the performance of MUSE automated analysis for most diagnoses, with some exceptions. The LIME explainability technique applied to CNNs highlights physiologically relevant ECG segments that contribute to the CNN's diagnoses.

The optical character recognition technique is used to convert information, mainly printed or handwritten text in paper materials, into an electronic format that the computers can edit. According to the literature, there are few competent OCR systems for recognizing multilingual characters in the form of Sinhala and English characters together. The lack of an appropriate technology to recognize multilingual text still remains as a problem that the current research community must address, and it has been designated as the key problem for this study. The main goal of this research is to develop a multilingual character recognition system that uses character image geometry features and Artificial Neural Networks to recognize printed Sinhala and English scripts together. It is intended that the solution would be improved to cover three Sri Lanka&#x2019;s most commonly spoken languages, with the addition of Tamil as a later upgrade. The primary technologies for this study were character geometry features and Artificial Neural Networks. At the moment almost an 85% of success rate has been achieved with a database containing around 800 images, which are divided into 46 characters (20 Sinhala and 26 English), and each character is represented in 20 different forms of character images. Recognition of text from printed bi-lingual documents is experimented by extracting individual character data from such printed text documents and feeding them to the system.

Image denoising faces significant challenges, arising from the sources of noise. Specifically, Gaussian, impulse, salt, pepper, and speckle noise are complicated sources of noise in imaging. Convolutional neural network (CNN) has increasingly received attention in image denoising task. Several CNN methods for denoising images have been studied. These methods used different datasets for evaluation. In this paper, we offer an elaborate study on different CNN techniques used in image denoising. Different CNN methods for image denoising were categorized and analyzed. Popular datasets used for evaluating CNN image denoising methods were investigated. Several CNN image denoising papers were selected for review and analysis. Motivations and principles of CNN methods were outlined. Some state-of-the-arts CNN image denoising methods were depicted in graphical forms, while other methods were elaborately explained. We proposed a review of image denoising with CNN. Previous and recent papers on image denoising with CNN were selected. Potential challenges and directions for future research were equally fully explicated.

The proliferation of social media platforms has led to a significant increase in the sharing of memes and other image-based content. However, some of these images may contain inappropriate or offensive text, which can be challenging to detect using traditional methods. Because the open-source existing solutions does not classify the text content in images. This project aims to develop a robust solution for detecting Not Safe for Work (NSFW) text content in images, with a particular focus on social media memes. The objectives of this project are to accurately extract text from images using optical character recognition (OCR) techniques, and to classify the extracted text as either NSFW or Safe for Work (SFW) using a fine-tuned natural language processing (NLP) model. The methodology employed in this project involves several key steps. First, the Keras OCR library is utilized to extract text from input images, as it has demonstrated superior performance compared to other OCR tools for social media images. Next, a preprocessing step is performed to align the extracted text in the correct sequence using coordinates, distance from the image origin, and the Pythagorean theorem. The pre-processed text is then passed to a fine-tuned BERT-BASE-UNCASED language model, which has been trained on a dataset of Reddit posts (NSFW and SFW) to classify the text as either NSFW or SFW. Key Words: OCR, Inappropriate, BERT, NSFW, SFW, Image, Text

Accurate and reliable detection of white matter hyperintensities and their volume quantification can provide valuable clinical information to assess neurologic disease progression. In this work, a stacked generalization ensemble of orthogonal 3D convolutional neural networks, StackGen-Net, is explored for improving automated detection of white matter hyperintensities in 3D T2-FLAIR images. Individual convolutional neural networks in StackGen-Net were trained on 2.5D patches from orthogonal reformatting of 3D-FLAIR (n = 21) to yield white matter hyperintensity posteriors. A meta convolutional neural network was trained to learn the functional mapping from orthogonal white matter hyperintensity posteriors to the final white matter hyperintensity prediction. The impact of training data and architecture choices on white matter hyperintensity segmentation performance was systematically evaluated on a test cohort (n = 9). The segmentation performance of StackGen-Net was compared with state-of-the-art convolutional neural network techniques on an independent test cohort from the Alzheimer's Disease Neuroimaging Initiative-3 (n = 20). StackGen-Net outperformed individual convolutional neural networks in the ensemble and their combination using averaging or majority voting. In a comparison with state-of-the-art white matter hyperintensity segmentation techniques, StackGen-Net achieved a significantly higher Dice score (0.76 [SD, 0.08], F1-lesion (0.74 [SD, 0.13]), and area under precision-recall curve (0.84 [SD, 0.09]), and the lowest absolute volume difference (13.3% [SD, 9.1%]). StackGen-Net performance in Dice scores (median = 0.74) did not significantly differ (P = .22) from interobserver (median = 0.73) variability between 2 experienced neuroradiologists. We found no significant difference (P = .15) in white matter hyperintensity lesion volumes from StackGen-Net predictions and ground truth annotations. A stacked generalization of convolutional neural networks, utilizing multiplanar lesion information using 2.5D spatial context, greatly improved the segmentation performance of StackGen-Net compared with traditional ensemble techniques and some state-of-the-art deep learning models for 3D-FLAIR.

this paper presents a literature review on OCR for different languages using convolutional neural network techniques. Optical Character Recognition is the process of converting an input text image into a machine encoded format. Different methods are used in OCR for different languages. The main steps of optical character recognition are pre-processing, segmentation and recognition. Recognizing handwritten text is harder than recognizing printed text. Convolutional Neural Network has shown remarkable improvement in recognizing characters of different languages. The novelty of the OCR is its robustness to image quality, image contrast, font style and font size. Common machine learning methods usually apply a combination of feature extractor and trainable classifier. The use of CNN leads to significant improvements across different machine-learning classification algorithms.