DOIONLINE 
DOIONLINE Publish In |
International Journal of Advance Computational Engineering and Networking (IJACEN)-IJACEN |
 Journal Home Volume Issue |
||||||||
Issue |
Volume-12,Issue-7 ( Jul, 2024 ) | |||||||||
Paper Title |
Face Liveness Detection Based on Features Fusion and Deep Learning Techniques | |||||||||
Author Name |
Jitendra Chautharia, Prasanth Bodepudi | |||||||||
Affilition |
||||||||||
Pages |
75-80 | |||||||||
Abstract |
Face liveness detection plays a crucial role in protecting facial recognition systems against spoofing attacks. In this paper, we present a comparative analysis of face liveness detection techniques, focusing on the fusion of feature extraction methods and the performance of traditional techniques versus deep learning models. We evaluate a range of feature extraction methods, including Local Binary Patterns (LBP), Histogram of Oriented Gradients (HOG), Gabor Wavelet and edge detection algorithms such as Step, Roberts, Sobel, and Laplacian. Additionally, we explore the effectiveness of combining these feature extraction methods and apply the Support Vector Machine (SVM) classifier to differentiate between real faces and spoofed images. Furthermore, we investigate the performance of pre-defined deep learning models, including Convolutional Neural Networks (CNN), ResNet50, MobileNetV3, and Inception, in face liveness detection. Through extensive experimentation on our dataset, we assess the strengths and limitations of each approach in terms of accuracy, robustness, and computational efficiency. Our findings provide valuable insights into the effectiveness of different techniques for face liveness detection and guiding the development of secure and reliable facial recognition systems. Keywords - Image classification, Feature extraction, Deep learning, Traditional methods, Comparative analysis, Convolutional Neural Networks, Support Vector Machine. | |||||||||
| View Paper | ||||||||||
