Digital image processing (DIP) and generative adversarial networks (GANS) techniques for improvement low-resolution face recognition
This research addresses the challenge of improving the accuracy of face recognition in lowresolution images using Digital Image Processing (DIP) and Generative Adversarial Networks (GANs). Recent advances in facial recognition have achieved high accuracy, although predominantly for high-resolution...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | en |
| Published: |
International Information and Engineering Technology Association
2024
|
| Online Access: | http://eprints.utem.edu.my/id/eprint/28505/2/00681301220241012571543.pdf http://eprints.utem.edu.my/id/eprint/28505/ https://www.iieta.org/pdf-viewer/18852 https://doi.org/10.18280/isi.290615 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This research addresses the challenge of improving the accuracy of face recognition in lowresolution images using Digital Image Processing (DIP) and Generative Adversarial
Networks (GANs). Recent advances in facial recognition have achieved high accuracy, although predominantly for high-resolution images. Low-resolution images, common in
surveillance and mobile devices, pose significant accuracy challenges. The proposed DIP+GAN method integrates image preprocessing techniques such as cropping, resizing,
normalization, and filtering with GANs to enhance low-resolution images. The study leverages the Georgia Tech Face Database for experiments and employs various DIP
techniques and GAN architecture. The results demonstrate improved facial recognition accuracy in low-resolution images and contribute significantly to the fields of digital image processing and artificial intelligence. This research highlights the importance of preprocessing in face recognition and the effectiveness of GANs in dealing with lowresolution images. |
|---|