Two Proposed Models for Face Recognition: Achieving High Accuracy and Speed with Artificial Intelligence

Engineering Technology & Applied Science Research, Journal Year: 2024, Volume and Issue: 14(2), P. 13706 - 13713

Published: April 2, 2024

In light of the development in computer science and modern technologies, impersonation crime rate has increased. Consequently, face recognition technology biometric systems have been employed for security purposes a variety applications including human-computer interaction, surveillance systems, etc. Building an advanced sophisticated model to tackle impersonation-related crimes is essential. This study proposes classification Machine Learning (ML) Deep (DL) models, utilizing Viola-Jones, Linear Discriminant Analysis (LDA), Mutual Information (MI), Variance (ANOVA) techniques. The two proposed facial are J48 with LDA feature extraction method as input, one-dimensional Convolutional Neural Network Hybrid Model (1D-CNNHM). MUCT database was considered training evaluation. performance, terms classification, reached 96.01% accuracy whereas DL that merged MI ANOVA 100% accuracy. Comparing models other works reflects they performing very well, high low processing time.

Language: Английский

---

A Partial Face Encryption in Real World Experiences Based on Features Extraction from Edge Detection

International Journal of Interactive Mobile Technologies (iJIM), Journal Year: 2023, Volume and Issue: 17(07), P. 69 - 81

Published: April 5, 2023

User confidentiality protection is concerning a topic in control and monitoring spaces. In image, user's faces security with compound information, abused situations, participation on global transmission media real-world experiences are extremely significant. For minifying the counting needs for vast size of image info time needful to be address computationally. consequently, partial encryption user-face picked. This study focuses large technique that designed encrypt face slightly. Primarily, dlib utilizing detection. Susan one top edge detectors valuable localization characteristics marked edges, used extract features vectors from user faces. Moreover, relevance suggested generating key led crucial role improvement by producing them as difficult intruders. According PSNR values, recommended algorithms provided an adequate outcome encryption, they had lower encrypting duration larger impact.

Language: Английский

---

A plaintext-related image encryption scheme based on key generation using generative adversarial networks

Multimedia Tools and Applications, Journal Year: 2024, Volume and Issue: unknown

Published: April 22, 2024

Language: Английский

---

Symmetric Keys for Lightweight Encryption Algorithms Using a Pre–Trained VGG16 Model

Telecom, Journal Year: 2024, Volume and Issue: 5(3), P. 892 - 906

Published: Sept. 3, 2024

The main challenge within lightweight cryptographic symmetric key systems is striking a delicate balance between security and efficiency. Consequently, the issue revolves around crafting schemes that are both robust enough to safeguard resource-constrained environments. This paper presents new method of making long keys for algorithms. A pre–trained convolutional neural network (CNN) model called visual geometry group 16 (VGG16) used take features from two images, turn them into binary strings, make strings equal by cutting down length shorter string, then use XOR images. depends on number in Compared other algorithms, we found this greatly decreases time required generate improves defense against brute force attacks creating exceptionally keys. successfully passed all 15 tests when evaluated using NIST SP 800-22 statistical test suite Basic Five Statistical Tests. To best our knowledge, first research explore generation encryption VGG16 model.

Language: Английский

---

Construction of Enterprise Economic Management Sharing Platform Based on Feature Recognition Grid System

Procedia Computer Science, Journal Year: 2024, Volume and Issue: 243, P. 593 - 600

Published: Jan. 1, 2024

Language: Английский

---