Histogram Of Oriented Gradients Research Articles

Driver inattention detection system using multi-task cascaded convolutional networks

Driver inattention has emerged as a critical concern impacting road safety, resulting in an alarming surge in accidents and fatalities. This research introduces a novel system for detecting inattention, structured across six levels: perception, facial feature extraction, tracking driver face, and driver secondary task using pre-trained deep learning models, inattention detection, risk estimation, and alert. The system is based on image processing captured from two strategically positioned cameras that simultaneously capture the driver’s activities while driving and their facial expressions. The second contribution concerns the driver facial features extraction using multi-task cascaded convolutional networks (MTCNN), and it is comparison with the histogram of gradient (HOG)-based frontal face detector, and haar feature based cascade classifier. The algorithms were compared based on their runtime efficiency, robustness in handling varying lighting conditions, and various head movements. The MTCNN achieves high performance, reaching accuracy levels ranging from 96.4% to 99.5% on two datasets including realistic driving scenarios: the DrivFace dataset and, the driver drowsiness dataset. The comparative analysis sheds light on the strengths and weaknesses of each algorithm, providing valuable insights for selecting the most suitable face detection algorithm to use in our system.

Traffic Sign Detection under Foggy Conditions using Machine Learning

Traffic sign detection is crucial for enhancing road safety, especially under adverse weather conditions like fog. This paper explores the challenges of detecting traffic signs in fog and presents an approach that incorporates dehazing techniques to improve detection accuracy. We implement a combination of YOLO (You Only Look Once) and HOG (Histogram of Oriented Gradients) algorithms, evaluating their performance on various datasets. Our findings indicate a significant improvement in detection rates under foggy conditions, suggesting that integrating dehazing techniques can enhance visibility and road safety.

An interpretable electrocardiogram-based model for predicting arrhythmia and ischemia in cardiovascular disease

IntroductionCardiovascular disease (CVD) is a leading cause of death and disability globally, with ischemia and arrhythmias being critical contributors. Ischemia, due to reduced myocardial blood flow, can lead to sudden cardiac death, while arrhythmias, marked by abnormal heart rhythms, are common in the elderly. Electrocardiography (ECG) is essential for the diagnosis of these conditions. This study aims to develop a clinically interpretable diagnostic framework for ischemia and arrhythmias using key ECG fiducial features. MethodsTo develop a robust ECG-based model for predicting cardiovascular diseases, we integrated data from three well-established ECG datasets: MIT-BIH Arrhythmia, European ST-T, and Fantasia. This aggregated dataset was employed to train multiple machine learning (ML) models aimed at automatically classifying heart conditions, including arrhythmia, ischemia, and healthy states. We designed a predictive framework utilizing boosting ML algorithms, enhanced by explainable artificial intelligence (XAI) techniques, to ensure high predictive performance in model interpretation. ResultsThe histogram gradient boosting classifier demonstrated superior classification performance, achieving an overall accuracy of 90 % in predicting heart disease based on ECG fiducial features. The model achieved area under the curve (AUC) scores of 0.99, 0.99, and 0.89 for the healthy, ischemic, and arrhythmic classes, respectively. XAI methods revealed that ECG fiducial features, such as the P-H, R-H, RR interval, QRS duration, QT interval, and ST segment, were significant diagnostic indicators for heart disease. ConclusionsThis study uses machine learning and XAI models to classify arrhythmia and ischemia from ECG data, enhancing interpretability clinical diagnostics for prevention and intervention to reduce disabilities.

Enhancing Structured Query Language Injection Detection with Trustworthy Ensemble Learning and Boosting Models Using Local Explanation Techniques

This paper presents a comparative analysis of several decision models for detecting Structured Query Language (SQL) injection attacks, which remain one of the most prevalent and serious security threats to web applications. SQL injection enables attackers to exploit databases, gain unauthorized access, and manipulate data. Traditional detection methods often struggle due to the constantly evolving nature of these attacks, the increasing complexity of modern web applications, and the lack of transparency in the decision-making processes of machine learning models. To address these challenges, we evaluated the performance of various models, including decision tree, random forest, XGBoost, AdaBoost, Gradient Boosting Decision Tree (GBDT), and Histogram Gradient Boosting Decision Tree (HGBDT), using a comprehensive SQL injection dataset. The primary motivation behind our approach is to leverage the strengths of ensemble learning and boosting techniques to enhance detection accuracy and robustness against SQL injection attacks. By systematically comparing these models, we aim to identify the most effective algorithms for SQL injection detection systems. Our experiments show that decision tree, random forest, and AdaBoost achieved the highest performance, with an accuracy of 99.50% and an F1 score of 99.33%. Additionally, we applied SHapley Additive exPlanations (SHAPs) and Local Interpretable Model-agnostic Explanations (LIMEs) for local explainability, illustrating how each model classifies normal and attack cases. This transparency enhances the trustworthiness of our approach to detecting SQL injection attacks. These findings highlight the potential of ensemble methods to provide reliable and efficient solutions for detecting SQL injection attacks, thereby improving the security of web applications.

Moving objects detection based on histogram of oriented gradient algorithm chip for hazy environment

<span>The most important aspects of computer vision are moving object detection (MOD) and tracking. Many signal-processing applications use regional image statistics. Compute-intensive video and image processing with low latency and high throughput is done with field programmable gate array (FPGA) image processing. Local image statistics are used for edge identification and filtering. The histogram of oriented gradients (HoG) algorithm extracts local shape characteristics by equalizing histograms. The objective of the work is to design the hardware chip of the algorithm and perform the simulation in the Xilinx ISE 14.7 simulation environment. The performance of the chip is evaluated in Modelsim 10.0 simulation software to check its feasibility. The performance of the chip design is estimated on Viretx-5 FPGA and compared with the MATLAB-2020 image processing tool-based response time. This form of tracking typically deals with identifying, anchoring, and tracking images and videos. A mask made from a cut-out of the object can then determine the plane's coordinates depending on its position. This type of object tracking is frequently utilized in the field of augmented reality (AR). The algorithm is most suited for object detection using hardware controllers in haze and foggy environments.</span>

Automated Bird Species Identification Through Machine Learning Techniques

The taxonomy of bird species is fundamental to ecological research, conservation efforts, and biodiversity monitoring. Traditional identification methods, which rely on field notes and visual assessments by trained ornithologists, are often labor-intensive, time-consuming, and prone to error. In recent years, machine learning algorithms and pre-trained models such as ResNet, Histogram of Oriented Gradients (HOG), and Scale-Invariant Feature Transform (SIFT) have shown significant promise in automating bird species classification. This study explores the application of these advanced models in identifying bird species from visual data, discussing key challenges, methodologies, and the potential to achieve high classification accuracy with reliable confidence scores. By leveraging deep learning techniques, we aim to enhance the precision and scalability of bird taxonomy, supporting more efficient ecological studies and conservation practices.

TinyML-Based Lightweight AI Healthcare Mobile Chatbot Deployment.

In healthcare applications, AI-driven innovations are set to revolutionise patient interactions and care, with the aim of improving patient satisfaction. Recent advancements in Artificial Intelligence have significantly affected nursing, assistive management, medical diagnoses, and other critical medical procedures. Many artificial intelligence (AI) solutions operate online, posing potential risks to patient data security. To address these security concerns and ensure swift operation, this study has developed a chatbot tailored for hospital environments, running on a local server, and utilising TinyML for processing patient data. Edge computing technology enables secure on-site data processing. The implementation includes patient identification using a Histogram of Gradient (HOG)-based classification, followed by basic patient care tasks, such as temperature measurement and demographic recording. The classification accuracy of patient detection was 95.8%. An autonomous temperature-sensing unit equipped with a medical-grade infrared temperature scanner detected and recorded patient temperature. Following the temperature assessment, the tinyML-powered chatbot engaged patients in a series of questions customised by doctors to train the model for diagnostic scenarios. Patients' responses, recorded as "yes" or "no", are stored and printed in their case sheet. The accuracy of the TinyML model is 95.3% and the on-device processing time is 217 ms. The implemented TinyML model uses only 8.8Kb RAM and 50.3Kb Flash memory, with a latency of only 4 ms. Each patient was assigned a unique ID, and their data were securely stored for further consultation and diagnosis via hospital management. This research demonstrates faster patient data recording and increased security compared to existing AI-based healthcare solutions, as all processes occur within the local host.

Investigation of Ways to Improve the HOG Method in the Classification of Histological Images by Machine Learning Methods

This study analyzes ways to improve the performance of the histogram of oriented gradients (HOG) method for histological image classification using machine learning methods. We propose applying HOG to each color channel of the image, which improves the extraction of texture features characteristic of histological data. A comparative analysis of the traditional HOG and the improved approach is conducted, including an experimental evaluation of their accuracy and processing time using SVM, RF, DT, KNN, and NB classifiers. The results show that the proposed improvements improve the classification accuracy, which makes the modified HOG method promising for application in biomedical analysis.

Enhancing multiclass brain tumor diagnosis using SVM and innovative feature extraction techniques

In the field of medical imaging, accurately classifying brain tumors remains a significant challenge because of the visual similarities among different tumor types. This research addresses the challenge of multiclass categorization by employing Support Vector Machine (SVM) as the core classification algorithm and analyzing its performance in conjunction with feature extraction techniques such as Histogram of Oriented Gradients (HOG) and Local Binary Pattern (LBP), as well as the dimensionality reduction technique, Principal Component Analysis (PCA). The study utilizes a dataset sourced from Kaggle, comprising MRI images classified into four classes, with images captured from various anatomical planes. Initially, the SVM model alone attained an accuracy(acc_val) of 86.57% on unseen test data, establishing a baseline for performance. To enhance this, PCA was incorporated for dimensionality reduction, which improved the acc_val to 94.20%, demonstrating the effectiveness of reducing feature dimensionality in mitigating overfitting and enhancing model generalization. Further performance gains were realized by applying feature extraction techniques—HOG and LBP—in conjunction with SVM, resulting in an acc_val of 95.95%. The most substantial improvement was observed when combining SVM with both HOG, LBP, and PCA, achieving an impressive acc_val of 96.03%, along with an F1 score(F1_val) of 96.00%, precision(prec_val) of 96.02%, and recall(rec_val) of 96.03%. This approach will not only improves categorization performance but also improves efficacy of computation, making it a robust and effective method for multiclass brain tumor prediction.

Research on the Registration of Aerial Images of Cyclobalanopsis Natural Forest Based on Optimized Fast Sample Consensus Point Matching with SIFT Features

The effective management and conservation of forest resources hinge on accurate monitoring. Nonetheless, individual remote-sensing images captured by low-altitude unmanned aerial vehicles (UAVs) fail to encapsulate the entirety of a forest’s characteristics. The application of image-stitching technology to high-resolution drone imagery facilitates a prompt evaluation of forest resources, encompassing quantity, quality, and spatial distribution. This study introduces an improved SIFT algorithm designed to tackle the challenges of low matching rates and prolonged registration times encountered with forest images characterized by dense textures. By implementing the SIFT-OCT (SIFT omitting the initial scale space) approach, the algorithm bypasses the initial scale space, thereby reducing the number of ineffective feature points and augmenting processing efficiency. To bolster the SIFT algorithm’s resilience against rotation and illumination variations, and to furnish supplementary information for registration even when fewer valid feature points are available, a gradient location and orientation histogram (GLOH) descriptor is integrated. For feature matching, the more computationally efficient Manhattan distance is utilized to filter feature points, which further optimizes efficiency. The fast sample consensus (FSC) algorithm is then applied to remove mismatched point pairs, thus refining registration accuracy. This research also investigates the influence of vegetation coverage and image overlap rates on the algorithm’s efficacy, using five sets of Cyclobalanopsis natural forest images. Experimental outcomes reveal that the proposed method significantly reduces registration time by an average of 3.66 times compared to that of SIFT, 1.71 times compared to that of SIFT-OCT, 5.67 times compared to that of PSO-SIFT, and 3.42 times compared to that of KAZE, demonstrating its superior performance.

Machine Learning in Agriculture: KNN-Based Classification of Fruits and Vegetables

In modern agriculture and the food industry, it is essential to classify fruits and vegetables accurately and efficiently to meet growing consumer demand and reduce post-harvest losses. Traditional manual methods are often labor-intensive and error prone, highlighting the need for automated solutions. This paper discusses the application of the k Nearest Neighbor (KNN) algorithm to fruit and vegetable recognition using image processing technology. In this study, image data sets are used for feature extraction using Directional Gradient Histogram (HOG) and dimensionality reduction using Principal Component Analysis (PCA). The accuracy of KNN model on verification set and test set reaches 97%, which proves its validity. Confusion matrix analysis and F1 score evaluation further revealed the strengths and areas of improvement of the model, particularly in distinguishing visually similar categories. The results show that the integration of artificial intelligence (especially KNN) offers great potential for the automation of agricultural classification tasks. Future studies could combine more advanced models, such as CNNS, with larger datasets to improve accuracy and robustness.

Research on the detection and recognition system of target vehicles based on fusion algorithm

<p>In modern society, the monitoring of vehicles is paramount for upholding traffic safety and order. This paper delves into a Graphical User Interface (GUI)-driven system for vehicle target detection and integration. This system seamlessly integrates three pivotal functionalities: vehicle type recognition, license plate recognition, and driver face recognition. It automates the vehicle inspection assessment process through an intuitive user interface. Specifically, for license plate recognition, the system leverages traditional computer vision techniques, including color and grayscale processing, radon transform, morphological operations, edge detection, character segmentation, and character recognition. The driver face recognition module incorporates methods such as image gray scaling, Gaussian filtering for denoising, face detection, and feature extraction. Meanwhile, vehicle type recognition is achieved by extracting Histogram of Oriented Gradients (HOG) features and utilizing a Support Vector Machine (SVM) classifier. Experimental findings highlight the system’s remarkable recognition accuracy, offering a highly efficient and dependable solution for the automated inspections of vehicles.</p>

Prediction of in-hospital mortality risk for patients with acute ST-elevation myocardial infarction after primary PCI based on predictors selected by GRACE score and two feature selection methods.

Accurate in-hospital mortality prediction following percutaneous coronary intervention (PCI) is crucial for clinical decision-making. Machine Learning (ML) and Data Mining methods have shown promise in improving medical prognosis accuracy. We analyzed a dataset of 4,677 patients from the Regional Vascular Center of Primorsky Regional Clinical Hospital No. 1 in Vladivostok, collected between 2015 and 2021. We utilized Extreme Gradient Boosting, Histogram Gradient Boosting, Light Gradient Boosting, and Stochastic Gradient Boosting for mortality risk prediction after primary PCI in patients with acute ST-elevation myocardial infarction. Model selection was performed using Monte Carlo Cross-validation. Feature selection was enhanced through Recursive Feature Elimination (RFE) and Shapley Additive Explanations (SHAP). We further developed hybrid models using Augmented Grey Wolf Optimizer (AGWO), Bald Eagle Search Optimization (BES), Golden Jackal Optimizer (GJO), and Puma Optimizer (PO), integrating features selected by these methods with the traditional GRACE score. The hybrid models demonstrated superior prediction accuracy. In scenario (1), utilizing GRACE scale features, the Light Gradient Boosting Machine (LGBM) and Extreme Gradient Boosting (XGB) models optimized with BES achieved Recall values of 0.944 and 0.954, respectively. In scenarios (2) and (3), employing SHAP and RFE-selected features, the LGB models attained Recall values of 0.963 and 0.977, while the XGB models achieved 0.978 and 0.99. The study indicates that ML models, particularly the XGB optimized with BES, can outperform the conventional GRACE score in predicting in-hospital mortality. The hybrid models' enhanced accuracy presents a significant step forward in risk assessment for patients post-PCI, offering a potential alternative to existing clinical tools. These findings underscore the potential of ML in optimizing patient care and outcomes in cardiovascular medicine.

Mathematical Evaluation of Deep Learning Architecture with Feature Fusion for Cervical Cancer Detection Classification

Cervical cancer is still one of the main reasons women die around the world, which shows how important it is to have accurate and quick ways to diagnose the disease. This study looks into how to improve the accuracy of finding cervical cancer using a deep learning-based classification system that combines features. In this mixed classic feature extraction methods, like Histogram of Oriented Gradients (HOG) and Scale-Invariant Feature Transform (SIFT), with deep learning architectures, like Graph Convolutional Networks (GCNs) and Capsule Networks. Each of these added its own benefits to the fusion model. Both HOG and SIFT are good at capturing important texture and structure features, even when the image conditions change. GCNs are used to learn relationship data between picture areas. This helps get a better sense of how things fit together in space, which is very important for medical imaging. Capsule Networks, on the other hand, let you understand the subtleties of hierarchical patterns in pictures of cervical cells. They do this by looking at spatial hierarchies and rotational changes that are very important for finding cancer. Our suggested fusion model combines these multi-level features, taking advantage of how the best features of deep learning and standard learning methods work together. Validation on standard cervical cancer image datasets showed that this fusion approach was much more accurate and precise at classifying than individual methods. Quantitative testing showed that the fusion model had higher sensitivity and specificity, which shows that it can reduce the number of fake negatives and positives, which is very important in clinical diagnosis. Our results show that the feature fusion model not only gives us a strong and accurate way to classify cervical cancer, but it also lays the groundwork for more uses in other complicated areas of medical imaging.

A New Computer-Aided Diagnosis System for Breast Cancer Detection from Thermograms Using Metaheuristic Algorithms and Explainable AI

Advances in the early detection of breast cancer and treatment improvements have significantly increased survival rates. Traditional screening methods, including mammography, MRI, ultrasound, and biopsies, while effective, often come with high costs and risks. Recently, thermal imaging has gained attention due to its minimal risks compared to mammography, although it is not widely adopted as a primary detection tool since it depends on identifying skin temperature changes and lesions. The advent of machine learning (ML) and deep learning (DL) has enhanced the effectiveness of breast cancer detection and diagnosis using this technology. In this study, a novel interpretable computer aided diagnosis (CAD) system for breast cancer detection is proposed, leveraging Explainable Artificial Intelligence (XAI) throughout its various phases. To achieve these goals, we proposed a new multi-objective optimization approach named the Hybrid Particle Swarm Optimization algorithm (HPSO) and Hybrid Spider Monkey Optimization algorithm (HSMO). These algorithms simultaneously combined the continuous and binary representations of PSO and SMO to effectively manage trade-offs between accuracy, feature selection, and hyperparameter tuning. We evaluated several CAD models and investigated the impact of handcrafted methods such as Local Binary Patterns (LBP), Histogram of Oriented Gradients (HOG), Gabor Filters, and Edge Detection. We further shed light on the effect of feature selection and optimization on feature attribution and model decision-making processes using the SHapley Additive exPlanations (SHAP) framework, with a particular emphasis on cancer classification using the DMR-IR dataset. The results of our experiments demonstrate in all trials that the performance of the model is improved. With HSMO, our models achieved an accuracy of 98.27% and F1-score of 98.15% while selecting only 25.78% of the HOG features. This approach not only boosts the performance of CAD models but also ensures comprehensive interpretability. This method emerges as a promising and transparent tool for early breast cancer diagnosis.

Bearing fault diagnosis based on novel hierarchical multiscale dispersion entropy in corresponding color block images

Abstract The rolling bearing is a critical component of mechanical equipment, and its failure can lead to serious consequences. In order to effectively extract fault features of rolling bearings and improve fault diagnosis performance, a fault diagnosis framework based on hierarchical multiscale dispersion entropy (HMDE) and improved histogram of oriented gradient (HOG) is proposed by combining entropy method with image recognition method. Firstly, the original vibration signal is subjected to moving average filtering to eliminate sudden noise and outliers. Then, HMDE is used for the extraction of fault features. HMDE can evaluate the complexity of the signal at different levels and scales, thereby extracting more comprehensive information. Based on HMDE, entropy color block (ECB) images are generated and the improved HOG of the images are extracted. Finally, Knearest neighbor (KNN) is used to classify the improved HOG features, completing the recognition of different working states of rolling bearings. The validity and robustness of the proposed fault diagnosis framework are proved by the verification experiments on the public bearing datasets in Case Western Reserve University and Southeast University.

Preservation of Cultural Heritage Buildings in Central Plains via Novel Detection Approach

Modern nations recognize cultural heritage as an expression of culture and variety. Conserving and repurposing historic buildings has only been more popular in the last decades. Nonetheless, a considerable portion of cultural legacy is afflicted by structural concerns that endanger the safety of buildings and people. Challenges include a scarcity of resources, whether financial, human, or material. This initiative aims to employ deep learning (DL) approaches to preserve cultural heritage buildings, particularly in poor nations where these buildings are still being maintained. To overcome these issues this study proposed a novel Flower Pollination improved Resnet (FP-IResNet) to detect preservation of cultural heritage buildings. The image data were collected from China's cultural heritage buildings. The data is preprocessed using normalization. Histogram of Oriented Gradients (HOG) using extract the features for preprocessed data. The proposed method is implemented using Python software. The findings reveal that the suggested obtained greater performance in the detection of cultural heritage buildings than other traditional algorithms. The suggested concept enables the computerized preservation of cultural heritage buildings, leading to improved accuracy and reduced individual fault. Performance measures show that the model is successful in correctly categorizing and detecting heritage buildings that require preservation, with high accuracy (96.82%), precision (97.21%), recall (97.58%), and an F1 score (93.58%). The research emphasizes how computerized techniques could enhance the precision and effectiveness of CH conservation initiatives

Oxygen Content Control in the Electroslag Remelting Process: An Incremental Learning Strategy Based on Optimized Wasserstein Generative Adversarial Network with Gradient Penalty Data Augmentation

Electroslag remelting (ESR) is essential for producing high‐end special steel, but its complex process and numerous influencing factors make quality control challenging. This study addresses oxygen content control during ESR using a big data machine learning approach. An incremental learning strategy is proposed based on an optimized Wasserstein generative adversarial network with gradient penalty (WGAN‐GP) for data enhancement, focusing on G20Cr2Ni4A bearing steel. The WGAN‐GP model enhances time‐series data and metadata, utilizing long short‐term memory networks, fully connected networks, and attention mechanisms. The effectiveness of data enhancement is verified using a deep neural network classifier and statistical methods. Data is divided into historical and data streams, with an incremental learning strategy based on histogram gradient boosting regression trees to prevent catastrophic forgetting and improve efficiency through knowledge distillation and real‐time hyperparameter adjustment. Results show that the data augmentation method significantly improves model generalization and accuracy in small sample metallurgy. The incremental learning strategy enhances prediction accuracy for oxygen content, contributing to better cleanliness quality of electroslag steel. This study offers a novel approach for addressing small sample challenges in metallurgical processes.

An adaptive learning based aberrance repressed multi-feature integrated correlation filter for Visual Object Tracking (VOT)

Target tracking via Correlation Filter (CF) is a hot research area of computer vision domain, and offers various credible benefits. Existing CF algorithms face challenges when there are target appearance variations due to background noise, scale and illumination changes, occlusion, and fast motion, which severely degrades the overall tracker performance. To get maximum benefits, an object tracker should perform well with the less computational burden in the presence of real time challenging situations. To address this issue, a novel visual object trackeris proposed based on multi feature fusion and adaptive learning technique with aberrance suppression. At first, multiple features i.e., Histogram of gradient (HOG), Color Naming (CN), saliency, and gray level intensities are combined using feature fusion technique. Further, based on the evaluation of final fused response map using Peak-to-Sidelobe Ratio (PSR), an adaptive learning strategy is integrated to improve the learning phase of tracker. Tracking results show that the proposed strategy beats the other modern CF trackers with Distance Precision (DP) scores of 88.2%, 85.9%, and 74.1% and 64.7% over OTB2013, OTB2015, and TempleColor128 and UAV123 datasets respectively.

An Automated Face Detection and Recognition for Class Attendance

Class attendance is a crucial indicator of students' seriousness towards learning. Many institutions continue to use manual methods, which are usually error-prone and unproductive. By leveraging computer vision algorithms, the system accurately captures and verifies the identity of students attending class. This paper aims to investigate and create an automated facial recognition system for classroom attendance to increase the precision and effectiveness of the attendance tracking system. To achieve this, we propose a system using computer vision technologies, namely Histogram of Oriented Gradients (HOG) with Support Vector Machine (SVM) for face detection and deep Convolutional Neural Networks (CNN) for face identification. The facial recognition system simplifies attendance recording, requiring participants to only gaze into the camera for the system to record their presence automatically. The system is rigorously tested and evaluated, and its accuracy is compared to our institution's current QR code attendance method. The study results reveal that the recommended approach is more accurate and competent than the existing procedures. The system allows for precise attendance records with real-time face detection and recognition capabilities. This technology ensures accurate and reliable attendance data, empowering organizations to make informed decisions, effectively manage resources, and provide a seamless experience for all students. In addition, a similar attendance system can be deployed for any event in an organization, thereby enhancing overall operational efficiency.