Search Results (22)

The automated digitization of handwritten electoral results is critical for ensuring transparency and speed in democratic processes. While recurrent sequence-to-sequence models (e.g., CRNN+CTC) achieve high accuracy, they inherently violate the strict latency constraints of high-throughput administrative environments. Conversely, standard lightweight CNNs exhibit suboptimal performance on the long-tail distribution of high-cardinality scenarios. To bridge this gap, this study reformulates sequence recognition into a latency-bound classification task. We propose a specialized Handwritten Digit Sequence Recognition (HDSR) framework for the Mexican Preliminary Election Results Program (PREP) based on a modified ResNet-18 architecture. The methodology introduces an asymmetric stride designed explicitly to preserve the 1:3 horizontal feature resolution of electoral tally sheets, integrating a lightweight Convolutional Block Attention Module (CBAM) in deep stages to refine classification across 1001 possible sequences. Leveraging a megadiverse dataset of 3.77 million real-world images, the model was trained using AdamW and label smoothing to mitigate human-induced label noise. Results demonstrate a global accuracy of 97.82% and a significant improvement in Macro-Precision (0.8878) for rare sequences. With an inference latency of 9.1 ms on standard CPU hardware, the proposed solution offers a scalable, high-confidence alternative that prioritizes spatial preservation and fail-controlled deployment. Full article

Vulnerable-code auditing is critical within the Software Development Life Cycle (SDLC) as timely remediation reduces exploitation risks from sensitive-data exposure to full system compromise. Although Automatic Program Repair (APR) has advanced through Machine Learning (ML), Deep Learning (DL), and Transformer-based generative models, current approaches remain limited by reduced language coverage, weak grounding in documented vulnerabilities, incomplete patch validation, and lack of explicit normative guidance for secure-coding compliance. This study introduces FocalVulNORM, a normatively conditioned Multi-Task repair model that generates a secure patch from vulnerable code while simultaneously verifying whether the vulnerable pattern is removed, the patch preserves observable behavior, and the repair aligns with the applicable security control. FocalVulNORM uses a shared Transformer encoder, a Focal Attention Layer, and four jointly optimized objectives: Repair, Detection, Normative Compliance, and Semantic Equivalence. The normative text is derived from CWE descriptions and SDLC secure-coding guides and standards, providing the security condition used to guide and evaluate each repair. Using this normative input, FocalVulNORM is trained on 11,503 vulnerability–patch pairs across seven programming languages and eight Common Weakness Enumeration (CWE) types. The model achieves CodeBLEU $=0.8247$ and Token-Level Accuracy (TLA) $=0.8976$ for Repair, $F1=0.9926$ for Detection, $F1=0.9820$ for Normative Compliance, and $F1=0.9921$ for Semantic Equivalence. The results indicate that FocalVulNORM extends APR beyond patch generation by attaching each repair to explicit evidence of vulnerability removal, behavioral preservation, and security-control alignment. Full article

During the early years of interaction between humans and computer systems, user authentication and identification was carried out with the support of knowledge-based factors (something the user knows: passwords, PINs, etc.) and tokens (something the user possesses: credentials, RFID cards, etc.) or a combination of both. In other words, the user presents a token and a password to the system in order to gain access. These solutions pose major challenges: Knowledge-based systems, which rely on secrets like passwords, are vulnerable to those secrets being guessed, shared, or forgotten. On the other hand, tokens are also vulnerable; some, despite implementing encryption, attract cyber attackers who can forge them, and users can share or lose them. In the search for more robust methods, the use of biometrics has been considered. Full article

Automated Program Repair (APR) is increasingly used for vulnerability patching, yet many existing methods focus primarily on syntactic similarity between vulnerable and fixed code, with limited guarantees of semantic correctness and limited alignment with security frameworks. This work presents a prototype-based deep metric learning method for vulnerability repair that integrates normative constraints from OWASP and NIST SSDF. The method combines embeddings of vulnerable code and CWE descriptions, refines category prototypes to improve separation among CWE types, and validates repairs against statement-level control requirements derived from the normative mapping. Experiments on 959 vulnerable–fixed pairs across Python, Java, C, and C++ covering 15 CWE categories achieved a Match Ratio of 88.95%, 0.81 compliance, and 0.84 consistency. Full article

The effectiveness of Machine Learning (ML)-based Network Intrusion Detection Systems (NIDS) is critically hampered by the scarcity of realistic and up-to-date malware traffic datasets. To address this gap, we present an automated platform for generating real-world malware traffic datasets. Our solution leverages a production-environment honeynet (T-Pot), deployed within a university network and segmented via a secure WireGuard VPN, to capture live attacks using high-interaction honeypots (Dionaea, Cowrie, ADBhoney). A fully automated pipeline handles traffic capture, transfer, filtering based on honeypot logs, and malware analysis (VirusTotal, VxAPI). The output is the IPN-UAN-23 dataset—a curated, labeled corpus of malicious network traffic. This platform functions as a vital automated security tool, providing the continuous stream of actionable intelligence required to develop and refine robust ML-based NIDS within a DevSecOps lifecycle. Full article

The digitization of electoral processes requires robust systems for processing handwritten numerical data from voting documents. This paper presents a convolutional neural network study for handwritten digit recognition in Mexico’s PREP (Programa de Resultados Electorales Preliminares) system. Rather than individual digit classification, we approach the problem as direct 1000-class classification, treating each three-digit combination as a single class to maximize accuracy and simplify inference. We evaluated eight CNN architectures including ResNet variants, MobileNetV3, ShuffleNetV2, and EfficientNet, with ResNet-18 emerging as optimal for balancing accuracy and computational efficiency under CPU-only deployment. To address dataset challenges including class imbalance and image artifacts, we developed a customized RandAugment strategy applying photometric and limited geometric transformations that preserve semantic integrity. Our methodology demonstrates feasibility of deploying robust digit recognition systems in resource-constrained electoral environments while maintaining high accuracy. The research provides a practical framework for automated electoral data processing adaptable to similar systems across Latin America. Full article

In recent years, the use of drones, also known as unmanned aerial vehicles (UAVs), has experienced a rapid increase due to their wide availability, compact size, low cost, and ease of operation. These devices have found applications in various areas, facilitating human work by covering large distances and operating in inaccessible or dangerous zones. However, their use has also been associated with malicious activities, such as property damage or threats to public security, which highlights the need to develop efficient and precise UAV detection systems. Although approaches based on neural networks have been proposed, they require large amounts of data for training and more computational resources for operation, which limits their applicability. In this study, we propose an alternative approach based on an analysis of audio features obtained through the fast Fourier transform (FFT) algorithm and classification using machine learning (ML) models. Our approach aims to detect the presence of drones using a minimal number of samples, meeting the requirements of efficiency, accuracy, robustness, low cost, and scalability necessary for a functional detection system. Full article

In Cyber Threat Intelligence, anticipating threat events and linking forecasts to standards-based mitigations is essential, yet many approaches rely on non-unified event representations within the analysis window, introducing bias and weakening tactical signal. In this manuscript, an end-to-end workflow is introduced that canonicalizes seven public threat feeds, constructs exogenous covariates, applies Elastic Net under Walk-Forward Cross-Validation (WFCV), and models continuous and intermittent series with SARIMAX and ADIDA estimators. Forecasts are consolidated into a fourteen-dimensional risk vector aligned with the MITRE ATT&CK framework taxonomy and translated into mitigations through a Retrieval-Augmented Generation (RAG) module that also consults the CISA Known Exploited Vulnerabilities catalogue. At a seven-day forecast horizon $h=7$ with weekly seasonality $m=7$, forecasting attains MAPE 12.0%, RMSE 6.8, and MAE 4.9. Mitigation retrieval, evaluated on 73 tactic-specific contextual queries from the test set, achieves 84.5% Exact Match and 91.3% Coverage. Full article

Instant messaging services are the main modern means of communication because they allow the exchange of messages between people anywhere and through many types of devices. Smishing involves sending text messages spoofing banks, government institutions, or companies in order to deceive. These messages often include malicious links that redirect users to fraudulent websites designed to steal personal information and commit financial fraud, identity theft, and extortion, among other crimes. Detecting smishing requires techniques to prevent access to dynamic links generated by cybercriminals to take control of devices or to consult blacklists of malicious links. Optical Character Recognition (OCR) recognizes text embedded in images without accessing links. This paper presents a conceptual model that uses OCR to extract text from messages suspected of smishing from a screenshot of a mobile device so that further processing can analyze whether it is smishing. Full article

In cybersecurity, identifying and addressing vulnerabilities in source code is essential for maintaining secure IT environments. Traditional static and dynamic analysis techniques, although widely used, often exhibit high false-positive rates, elevated costs, and limited interpretability. Machine Learning (ML)-based approaches aim to overcome these limitations but encounter challenges related to scalability and adaptability due to their reliance on large labeled datasets and their limited alignment with the requirements of secure development teams. These factors hinder their ability to adapt to rapidly evolving software environments. This study proposes an approach that integrates Prototype-Based Model-Agnostic Meta-Learning(Proto-MAML) with a Question-Answer (QA) framework that leverages the Bidirectional Encoder Representations from Transformers (BERT) model. By employing Few-Shot Learning (FSL), Proto-MAML identifies and mitigates vulnerabilities with minimal data requirements, aligning with the principles of the Secure Development Lifecycle (SDLC) and Development, Security, and Operations (DevSecOps). The QA framework allows developers to query vulnerabilities and receive precise, actionable insights, enhancing its applicability in dynamic environments that require frequent updates and real-time analysis. The model outputs are interpretable, promoting greater transparency in code review processes and enabling efficient resolution of emerging vulnerabilities. Proto-MAML demonstrates strong performance across multiple programming languages, achieving an average precision of $98.49\%$, recall of $98.54\%$, F1-score of $98.78\%$, and exact match rate of $98.78\%$ in PHP, Java, C, and C++. Full article

Conducting penetration testing (pentesting) in cybersecurity is a crucial turning point for identifying vulnerabilities within the framework of Information Technology (IT), where real malicious offensive behavior is simulated to identify potential weaknesses and strengthen preventive controls. Given the complexity of the tests, time constraints, and the specialized level of expertise required for pentesting, analysis and exploitation tools are commonly used. Although useful, these tools often introduce uncertainty in findings, resulting in high rates of false positives. To enhance the effectiveness of these tests, Machine Learning (ML) has been integrated, showing significant potential for identifying anomalies across various security areas through detailed detection of underlying malicious patterns. However, pentesting environments are unpredictable and intricate, requiring analysts to make extensive efforts to understand, explore, and exploit them. This study considers these challenges, proposing a recommendation system based on a context-rich, vocabulary-aware transformer capable of processing questions related to the target environment and offering responses based on necessary pentest batteries evaluated by a Reinforcement Learning (RL) estimator. This RL component assesses optimal attack strategies based on previously learned data and dynamically explores additional attack vectors. The system achieved an F1 score and an Exact Match rate over $97.0$%, demonstrating its accuracy and effectiveness in selecting relevant pentesting strategies. Full article

The rapid evolution of the Internet of Everything (IoE) has significantly enhanced global connectivity and multimedia content sharing, simultaneously escalating the unauthorized distribution of multimedia content, posing risks to intellectual property rights. In 2022 alone, about 130 billion accesses to potentially non-compliant websites were recorded, underscoring the challenges for industries reliant on copyright-protected assets. Amidst prevailing uncertainties and the need for technical and AI-integrated solutions, this study introduces two pivotal contributions. First, it establishes a novel taxonomy aimed at safeguarding and identifying IoE-based content infringements. Second, it proposes an innovative architecture combining IoE components with automated sensors to compile a dataset reflective of potential copyright breaches. This dataset is analyzed using a Bidirectional Encoder Representations from Transformers-based advanced Natural Language Processing (NLP) algorithm, further fine-tuned by a dense neural network (DNN), achieving a remarkable 98.71% accuracy in pinpointing websites that violate copyright. Full article

Facial recognition is fundamental for a wide variety of security systems operating in real-time applications. Recently, several deep neural networks algorithms have been developed to achieve state-of-the-art performance on this task. The present work was conceived due to the need for an efficient and low-cost processing system, so a real-time facial recognition system was proposed using a combination of deep learning algorithms like FaceNet and some traditional classifiers like SVM, KNN, and RF using moderate hardware to operate in an unconstrained environment. Generally, a facial recognition system involves two main tasks: face detection and recognition. The proposed scheme uses the YOLO-Face method for the face detection task which is a high-speed real-time detector based on YOLOv3, while, for the recognition stage, a combination of FaceNet with a supervised learning algorithm, such as the support vector machine (SVM), is proposed for classification. Extensive experiments on unconstrained datasets demonstrate that YOLO-Face provides better performance when the face under an analysis presents partial occlusion and pose variations; besides that, it can detect small faces. The face detector was able to achieve an accuracy of over 89.6% using the Honda/UCSD dataset which runs at 26 FPS with darknet-53 to VGA-resolution images for classification tasks. The experimental results have demonstrated that the FaceNet+SVM model was able to achieve an accuracy of 99.7% using the LFW dataset. On the same dataset, FaceNet+KNN and FaceNet+RF achieve 99.5% and 85.1%, respectively; on the other hand, the FaceNet was able to achieve 99.6%. Finally, the proposed system provides a recognition accuracy of 99.1% and 49 ms runtime when both the face detection and classifications stages operate together. Full article

This paper proposes the use of the FASSD-Net model for semantic segmentation of human silhouettes, these silhouettes can later be used in various applications that require specific characteristics of human interaction observed in video sequences for the understanding of human activities or for human identification. These applications are classified as high-level task semantic understanding. Since semantic segmentation is presented as one solution for human silhouette extraction, it is concluded that convolutional neural networks (CNN) have a clear advantage over traditional methods for computer vision, based on their ability to learn the representations of appropriate characteristics for the task of segmentation. In this work, the FASSD-Net model is used as a novel proposal that promises real-time segmentation in high-resolution images exceeding 20 FPS. To evaluate the proposed scheme, we use the Cityscapes database, which consists of sundry scenarios that represent human interaction with its environment (these scenarios show the semantic segmentation of people, difficult to solve, that favors the evaluation of our proposal), To adapt the FASSD-Net model to human silhouette semantic segmentation, the indexes of the 19 classes traditionally proposed for Cityscapes were modified, leaving only two labels: One for the class of interest labeled as person and one for the background. The Cityscapes database includes the category “human” composed for “rider” and “person” classes, in which the rider class contains incomplete human silhouettes due to self-occlusions for the activity or transport used. For this reason, we only train the model using the person class rather than human category. The implementation of the FASSD-Net model with only two classes shows promising results in both a qualitative and quantitative manner for the segmentation of human silhouettes. Full article

This paper proposes a teleophthalmology support system in which we use algorithms of object detection and semantic segmentation, such as faster region-based CNN (FR-CNN) and SegNet, based on several CNN architectures such as: Vgg16, MobileNet, AlexNet, etc. These are used to segment and analyze the principal anatomical elements, such as optic disc (OD), region of interest (ROI) composed by the macular region, real retinal region, and vessels. Unlike the conventional retinal image quality assessment system, the proposed system provides some possible reasons about the low-quality image to support the operator of an ophthalmoscope and patient to acquire and transmit a better-quality image to central eye hospital for its diagnosis. The proposed system consists of four steps: OD detection, OD quality analysis, obstruction detection of the region of interest (ROI), and vessel segmentation. For the OD detection, artefacts and vessel segmentation, the FR-CNN and SegNet are used, while for the OD quality analysis, we use transfer learning. The proposed system provides accuracies of 0.93 for the OD detection, 0.86 for OD image quality, 1.0 for artefact detection, and 0.98 for vessel segmentation. As the global performance metric, the kappa-based agreement score between ophthalmologist and the proposed system is calculated, which is higher than the score between ophthalmologist and general practitioner. Full article