An Ensemble Kernelized-based Approach for Precise Emotion Recognition in Depressed People DOI Open Access

B. Sahoo,

Arpita Gupta

Engineering Technology & Applied Science Research, Год журнала: 2024, Номер 14(6), С. 18873 - 18882

Опубликована: Дек. 2, 2024

As the COVID-19 pandemic created serious challenges for mental health worldwide, with a noticeable increase in depression cases, it has become important to quickly and accurately assess emotional states. Facial expression recognition technology is key tool this task. To address need, study proposes new approach emotion using Ensemble Kernelized Learning System (EKLS). Nonverbal cues, such as facial expressions, are crucial showing This uses Extended Cohn-Kanade (CK+) dataset, which was enhanced images videos from era related depression. Each of these manually labeled corresponding emotions, creating strong dataset training testing proposed model. feature detection techniques were used along measurements aid recognition. EKLS flexible machine-learning framework that combines different techniques, including Support Vector Machines (SVMs), Self-Organizing Maps (SOMs), kernel methods, Random Forest (RF), Gradient Boosting (GB). The ensemble model thoroughly trained fine-tuned ensure high accuracy consistency. powerful real-time both videos, achieving an impressive 99.82%. offers practical effective makes significant contribution field.

Язык: Английский

An Enhanced Document Source Identification System for Printer Forensic Applications based on the Boosted Quantum KNN Classifier DOI Open Access

Shahlaa Mashhadani,

Wisal Hashim Abdulsalam,

Iptehaj Alhakam

и другие.

Engineering Technology & Applied Science Research, Год журнала: 2025, Номер 15(1), С. 19983 - 19991

Опубликована: Фев. 2, 2025

Document source identification in printer forensics involves determining the origin of a printed document based on characteristics such as model, serial number, defects, or unique printing artifacts. This process is crucial forensic investigations, particularly cases involving counterfeit documents unauthorized printing. However, consistent pattern across various types remains challenging, especially when efforts are made to alter printer-generated Machine learning models often used these tasks, but selecting discriminative features while minimizing noise essential. Traditional KNN classifiers require careful selection distance metrics capture relevant effectively. study proposes leveraging quantum-inspired computing improve for identification, offering better accuracy even with noisy variable conditions. The proposed approach uses Gray Level Co-occurrence Matrix (GLCM) feature extraction, which resilient changes rotation and scale, making it well-suited texture analysis. Experimental results show that classifier captures subtle artifacts, leading improved classification despite variability.

Язык: Английский

Процитировано

0
Accuracy of web-based automated versus digital manual cephalometric landmark identification DOI
Mais Sadek,

Omar Alaskari,

Ahmad Hamdan

и другие.

Clinical Oral Investigations, Год журнала: 2024, Номер 28(11)

Опубликована: Ноя. 1, 2024

Язык: Английский

Процитировано

4
DETECTION OF KERATOCONUS DISEASE DEPENDING ON CORNEAL TOPOGRAPHY USING DEEP LEARNING DOI Creative Commons

AM Kamrul Hasan,

Mahdi Mazinani

Kufa Journal of Engineering, Год журнала: 2025, Номер 16(1), С. 463 - 478

Опубликована: Фев. 4, 2025

Keratoconus is a disease that ML has contributed much in its diagnosis and management. It not widely prevalent disease, with research gap caused by the absence of standardized datasets for model training evaluation. This work presents novel dataset, which strengthens CNN model's resilience creates standards assessing keratoconus diagnostic techniques. The depends on data patients examined at Jenna Ophthalmic Center Baghdad. proposed system works three stages: pre-processing, feature extraction, classification machine learning algorithms including NB, KNN, ADA, DT, deep learning. pre-processing stage involves cropping images to retain relevant maps, were subjected contrast enhancement improve image quality. pre-processed then fed into Machine Learning(ML) Convolutional Neural Network(CNN) models, four corneal maps analyzed. precision method was quantified, yielding score 0.79 AdaBoost algorithm an impressive 0.99 suggested exemplifying high accuracy ability surpass all approaches. Applying PCA extraction before utilizing tradition helps achieving high-accuracy results.

Язык: Английский

Процитировано

0
Optimized hybrid SVM-RF multi-biometric framework for enhanced authentication using fingerprint, iris, and face recognition DOI Creative Commons

Sonal Sonal,

Ajit Singh, Chander Kant

и другие.

PeerJ Computer Science, Год журнала: 2025, Номер 11, С. e2699 - e2699

Опубликована: Фев. 17, 2025

This article introduces a hybrid multi-biometric system incorporating fingerprint, face, and iris recognition to enhance individual authentication. The addresses limitations of uni-modal approaches by combining multiple biometric modalities, exhibiting superior performance heightened security in practical scenarios, making it more dependable resilient for real-world applications. integration support vector machine (SVM) random forest (RF) classifiers, along with optimization techniques like bacterial foraging (BFO) genetic algorithms (GA), improves efficiency robustness. Additionally, integrating feature-level fusion utilizing methods such as Gabor filters feature extraction enhances overall the model. demonstrates accuracy reliability, suitable applications requiring secure identification solutions.

Язык: Английский

Процитировано

0
Deep Learning and Fusion Mechanism-based Multimodal Fake News Detection Methodologies: A Review DOI Open Access
Iman Qays Abduljaleel,

Israa Hadi Ali

Engineering Technology & Applied Science Research, Год журнала: 2024, Номер 14(4), С. 15665 - 15675

Опубликована: Авг. 2, 2024

Today, detecting fake news has become challenging as anyone can interact by freely sending or receiving electronic information. Deep learning processes to detect multimodal have achieved great success. However, these methods easily fuse information from different modality sources, such concatenation and element-wise product, without considering how each affects the other, resulting in low accuracy. This study presents a focused survey on use of deep approaches visual textual various social networks 2019 2024. Several relevant factors are discussed, including a) detection stage, which involves algorithms, b) for analyzing data types, c) choosing best fusion mechanism combine multiple sources. delves into existing constraints previous studies provide future tips addressing open challenges problems.

Язык: Английский

Процитировано

2
Emotional Facial Expression Detection using YOLOv8 DOI Open Access
Aadil Alshammari,

Muteb E. Alshammari

Engineering Technology & Applied Science Research, Год журнала: 2024, Номер 14(5), С. 16619 - 16623

Опубликована: Окт. 9, 2024

Emotional facial expression detection is a critical component with applications ranging from human-computer interaction to psychological research. This study presents an approach emotion using the state-of-the-art YOLOv8 framework, Convolutional Neural Network (CNN) designed for object tasks. utilizes dataset comprising 2,353 images categorized into seven distinct emotional expressions: anger, contempt, disgust, fear, happiness, sadness, and surprise. The findings suggest that framework promising tool detection, potential further enhancement through augmentation. research demonstrates feasibility effectiveness of advanced CNN architectures recognition

Язык: Английский

Процитировано

2
Safeguarding Identities with GAN-based Face Anonymization DOI Open Access
Mahmoud Ahmad Al‐Khasawneh, Marwan Mahmoud

Engineering Technology & Applied Science Research, Год журнала: 2024, Номер 14(4), С. 15581 - 15589

Опубликована: Авг. 2, 2024

Effective anonymous facial registration techniques are critical to address privacy concerns arising from recognition technology. This study presents an intelligent anonymity platform that incorporates blockchain with advanced and uses a CIAGAN-powered approach. solution addresses the immediate need for in The proposed system anonymously generate highly realistic effective images. widespread use of systems places greater emphasis on concerns, emphasizing strong enrollment mechanisms. CIAGAN this challenge images while preserving important attributes. Blockchain storage ensures data integrity security maintained. process begins detailed image preprocessing steps improve quality eliminate unwanted noise. can face attributes complicate specific objects. A dataset 202,599 was used. Performance metrics such as PSNR SSIM indicate uniformity. obtained 35.0516, indicating unique anonymization process.

Язык: Английский

Процитировано

1
An Ensemble Kernelized-based Approach for Precise Emotion Recognition in Depressed People DOI Open Access

B. Sahoo,

Arpita Gupta

Engineering Technology & Applied Science Research, Год журнала: 2024, Номер 14(6), С. 18873 - 18882

Опубликована: Дек. 2, 2024

As the COVID-19 pandemic created serious challenges for mental health worldwide, with a noticeable increase in depression cases, it has become important to quickly and accurately assess emotional states. Facial expression recognition technology is key tool this task. To address need, study proposes new approach emotion using Ensemble Kernelized Learning System (EKLS). Nonverbal cues, such as facial expressions, are crucial showing This uses Extended Cohn-Kanade (CK+) dataset, which was enhanced images videos from era related depression. Each of these manually labeled corresponding emotions, creating strong dataset training testing proposed model. feature detection techniques were used along measurements aid recognition. EKLS flexible machine-learning framework that combines different techniques, including Support Vector Machines (SVMs), Self-Organizing Maps (SOMs), kernel methods, Random Forest (RF), Gradient Boosting (GB). The ensemble model thoroughly trained fine-tuned ensure high accuracy consistency. powerful real-time both videos, achieving an impressive 99.82%. offers practical effective makes significant contribution field.

Язык: Английский

Процитировано

0