Deep Learning Approach to Recyclable Products Classification: Towards Sustainable Waste Management

Sustainability, Journal Year: 2023, Volume and Issue: 15(14), P. 11138 - 11138

Published: July 17, 2023

Effective waste management and recycling are essential for sustainable development environmental conservation. It is a global issue around the globe emerging in Saudi Arabia. The traditional approach to sorting relies on manual labor, which both time-consuming, inefficient, prone errors. Nonetheless, rapid advancement of computer vision techniques has paved way automating garbage classification, resulting enhanced efficiency, feasibility, management. In this regard, study, comprehensive investigation classification using state-of-the-art algorithm, such as Convolutional Neural Network (CNN), well pre-trained models DenseNet169, MobileNetV2, ResNet50V2 been presented. As an outcome CNN model achieved accuracy 88.52%, while ResNet50V2, 94.40%, 97.60%, 98.95% accuracies, respectively. That considerable contrast studies literature. proposed study potential contribution facilitating effective system more greener future. Consequently, it may alleviate burden reduce human error, encourage practices, ultimately promoting

Language: Английский

---

Joint Diagnosis of Pneumonia, COVID-19, and Tuberculosis from Chest X-ray Images: A Deep Learning Approach

Diagnostics, Journal Year: 2023, Volume and Issue: 13(15), P. 2562 - 2562

Published: Aug. 1, 2023

Pneumonia, COVID-19, and tuberculosis are some of the most fatal common lung diseases in current era. Several approaches have been proposed literature for diagnosis individual diseases, since each requires a different feature set altogether, but few studies joint diagnosis. A patient being diagnosed with one disease as negative may be suffering from other disease, vice versa. However, said related to lungs, there might likelihood more than present same patient. In this study, deep learning model that is able detect mentioned chest X-ray images patients proposed. To evaluate performance model, multiple public datasets obtained Kaggle. Consequently, achieved 98.72% accuracy all classes general recall score 99.66% 99.35% No-findings, 98.10% Tuberculosis, 96.27% respectively. Furthermore, was tested using unseen data augmented dataset proven better state-of-the-art terms metrics.

Language: Английский

---

A Deep-Learning Approach to Driver Drowsiness Detection

Safety, Journal Year: 2023, Volume and Issue: 9(3), P. 65 - 65

Published: Sept. 13, 2023

Drowsy driving is a widespread cause of traffic accidents, especially on highways. It has become an essential task to seek understanding the situation in order be able take immediate remedial actions detect driver drowsiness and enhance road safety. To address issue safety, proposed model offers method for evaluating level fatigue based changes driver’s eyeball movement using convolutional neural network (CNN). Further, with help CNN VGG16 models, facial sleepiness expressions were detected classified into four categories (open, closed, yawning, no yawning). Subsequently, dataset 2900 images eye conditions associated was used test which include different range features such as gender, age, head position, illumination. The results devolved models show high degree accountability, whereas achieved accuracy rate 97%, precision 99%, recall F-score values 99%. reached 74%. This considerable contrast between state-of-the-art methods literature similar problems.

Language: Английский

---

Optimization-driven artificial intelligence-enhanced municipal waste classification system for disaster waste management

Engineering Applications of Artificial Intelligence, Journal Year: 2024, Volume and Issue: 133, P. 108614 - 108614

Published: May 30, 2024

This research addresses the critical challenge of disaster waste management, a growing concern exacerbated by increasing frequency and intensity natural disasters like flooding. Traditional systems often struggle with volume heterogeneity waste, highlighting need for innovative solutions. In this study, we present novel classification model integrating advanced artificial intelligence (AI) optimization techniques to streamline categorization in post-disaster environments. Our approach leverages dual ensemble deep learning framework. The first combines various image-segmentation methods, while second integrates outputs from diverse convolutional neural network architectures. A modified multiple system serves as decision fusion strategy, enhancing accuracy at both points. We rigorously evaluated our using three datasets: "TrashNet" dataset benchmarking against existing well two meticulously curated, real-world datasets collected flood-affected areas Thailand. results demonstrate that method outperforms algorithms VGG19, YoloV5, InceptionV3 general solid classification, achieving an average improvement 11.18%. Regarding specifically, achieves 96.48% 96.49% on curated datasets, consistently outperforming ResNet-101, DenseNet-121, 3.47%. These findings potential AI-enhanced revolutionize management practices. Thus, advocate such technologies into municipal policies enhance resilience optimize responses. Future will explore scaling types incorporating real-time data adaptable strategies.

Language: Английский

---

Transfer Learning for Enhancing Computer Vision

Lecture notes in networks and systems, Journal Year: 2025, Volume and Issue: unknown, P. 773 - 786

Published: Jan. 1, 2025

Language: Английский

---

Oil and Gas Pipelines Leakage Detection Approaches: A Systematic Review of Literature

International Journal of Safety and Security Engineering, Journal Year: 2024, Volume and Issue: 14(3), P. 773 - 786

Published: June 24, 2024

In terms of its significance, the oil & gas industry is ranked among top global industries.Like any other industry, it also faces various problems, such as leakage and pipelines.The detection in pipelines essential for an or plant to operate properly maintain environmental safety well minimize supply-chain losses.The undergoing study systematically reviews literature comprising more than a decade (2010-2021) span summarize systems, methods techniques used pipeline detection.Likewise, this paper investigates effective low-cost systems with their pros cons.The existing are classified into three categories based on technical characteristics, named hardware-based (where some hardware deployed monitoring leakage), software-based software intelligent predictive algorithm detection) techniques.Each technique was reviewed according datasets used, preprocessing (mainly that imagery image largely like enhancement, denoising filtering), investigated classifiers' efficiencies, results, limitations.A comparative analysis conducted help determine which technology best given operational environment, software, hardware, hybrid.Further, highlights gaps research unresolved concerns regarding development dependable leak suggests possible directions mitigate it.

Language: Английский

---

Blockchain Empowered Interoperable Framework for Smart Healthcare

Mathematical Modelling and Engineering Problems, Journal Year: 2024, Volume and Issue: 11(5), P. 1330 - 1340

Published: May 30, 2024

In the past, healthcare industry used paper-based systems to manage and store medical records.However, these are vulnerable data breaches, loss, errors.To overcome issues, a research study has been conducted create safe efficient Electronic Data Interchange (EDI) system for using blockchain technology.The utilized various tools methods including Python as programming language implement environment, pyQT5 library graphical user interface (GUI), MySQL database management repository Health Records (EHR) with DBeaver, cross-platform tool management.The work involves development of blockchain-based smart contract storage, exchange, retrieval EHR.Additionally, application based on is created provide users friendly GUI.The proposed provides secure platform storing managing EHR well enabling EDI among stakeholders like practices, doctors, labs, pharmacies.Furthermore, scalable user-friendly, includes features patient visits, history, appointment scheduling.Blockchain technology ensures integrity, EDI, confidentiality, while user-friendly enhances experience compared existing standards health level 7 (HL7).

Language: Английский

---

Early Detection of Diabetic Retinopathy Utilizing Advanced Fuzzy Logic Techniques

Mathematical Modelling and Engineering Problems, Journal Year: 2023, Volume and Issue: 10(6), P. 2086 - 2094

Published: Dec. 21, 2023

The escalating prevalence of diabetes globally, exacerbated by lifestyle changes postpandemic-including increased screen time, sedentary behavior, and remote workhas consequently driven a surge in associated complications, notably, Diabetic Retinopathy (DR).This ocular complication presents pressing concern due to its potential precipitate irreversible vision loss.Consequently, the necessity for timely accurate DR detection is paramount, especially circumstances where conventional diagnostic approaches are either challenging or financially prohibitive.Capitalizing on prowess fuzzy logic managing uncertainties, this study introduces an innovative application Extended Fuzzy Logic early-stage DR.Rather than focusing solely overt symptoms, approach discerns subtle similarities retinal irregularities between diabetic patients non-diabetic individuals.To quantify these similarities, 'f-validity' value was computed based risk factors which were subsequently transformed into membership function values.The aggregation values facilitated Ordered Weighted Averaging (OWA) operator.The experimental outcomes align satisfactorily with expert anticipations, boasting accuracy 90%, precision 92.2%, sensitivity 75%.These results, when juxtaposed against contemporary studies field, underscore promise scheme advancing early diagnostics DR.The thus proposes solution that leverages power address burgeoning challenge DR.

Language: Английский

---

Predicting Global Energy Consumption Through Data Mining Techniques

International Journal of Design & Nature and Ecodynamics, Journal Year: 2024, Volume and Issue: 19(2), P. 397 - 406

Published: April 25, 2024

With the explosion of global population and technological progress, electricity demand has skyrocketed.To ensure a consistent flow power, it's essential to accurately predict energy usage ahead time.Failure do so could lead potential outages disrupt our daily lives.This research reviews previous in field using data mining techniques analyze consumption data, optimize performance buildings, various industries.The study also aims uncover patterns, correlations, rules worldwide techniques.The analysis is performed techniques, such as simple K-Means Expectation Maximization (EM).This selection based on their prominent applications for similar problems literature.The EM algorithms showed successful outcomes dataset, which evident clustering plots.Further, Hierarchical Clustering algorithm was not up desired standard.This probably due nature available dataset.These will provide valuable resource decision-makers stakeholders sector, it deeper understanding patterns trends.This sustainable future.

Language: Английский

---

Rule-Based Information Extraction from Multi-format Resumes for Automated Classification

Mathematical Modelling and Engineering Problems, Journal Year: 2024, Volume and Issue: 11(4), P. 1044 - 1052

Published: April 26, 2024

Nowadays, with the expansion of Internet, a lot people publish their resumes on internet and social media networks.Large companies receive hundreds per day, which comes in several formats such as Joint Photographic Experts Group (JPG), Portable Document Format (PDF) Word files.Therefore, information extraction from can be applied automatically by methods.In this research, important details that are taken are: name, date birth, email, phone number, GPA, gender, nationality, address.The private dataset used is different sources including open source well personally annotated.The processes for have been performed phases as: pre-processing, converting files into PDF rule-based method to extract eight elements resumes.To carry out experiment, Python language used, particularly spacy library word2vec technique.Consequently, experimental results demonstrate testing phase achieved 96.4% precision quite considerable contrast techniques literature.The scheme then extended classify resume based extracted fields exhibited classification accuracy, precision, recall F1-score 98.02%, 98.01%, 98% 98%, respectively.

Language: Английский

---