IEMEK Journal of Embedded Systems and Applications (대한임베디드공학회논문지)

Institute of Embedded Engineering of Korea (대한임베디드공학회)
권호 표지
DOI QR코드

DOI QR Code

Deep Learning-Based Face Recognition through Low-Light Enhancement

딥러닝 기반 저조도 향상 기술을 활용한 얼굴 인식 성능 개선

  • Changwoo Baek (Pusan National University) ;
  • Kyeongbo Kong (Pusan National University)
  • 백창우 ;
  • 공경보
  • Received : 2024.06.30
  • Accepted : 2024.08.28
  • Published : 2024.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

This study explores enhancing facial recognition performance in low-light environments using deep learning-based low-light enhancement techniques. Facial recognition technology is widely used in edge devices like smartphones, smart home devices, and security systems, but low-light conditions reduce accuracy due to degraded image quality and increased noise. We reviewed the latest techniques, including Zero-DCE, Zero-DCE++, and SCI (Self-Calibrated Illumination), and applied them as preprocessing steps in facial recognition on edge devices. Using the K-face dataset, experiments on the Qualcomm QRB5165 platform showed significant improvements in F1 SCORE from 0.57 to 0.833 with SCI. Processing times were 0.15ms for SCI, 0.4ms for Zero-DCE, and 0.7ms for Zero-DCE++, all much shorter than the facial recognition model MobileFaceNet's 5ms. These results indicate that these techniques can be effectively used in resource-limited edge devices, enhancing facial recognition in low-light conditions for various applications.

Keywords

Acknowledgement

이 과제는 부산대학교 기본연구지원사업 (2년)에 의하여 연구되었음.

References

  1. C. Guo, C. Li, J. Guo, C. C. Loy, J. Hou, S. Kwong, R. Cong, "Zero-reference Deep Curve Estimation for Low-light Image Enhancement." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 1780-1789, 2020.
  2. C. Li, C. Guo, C. C. Loy, "Learning to Enhance Low-light Image Via Zero-reference Deep Curve Estimation." IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 44, No. 8, pp. 4225-4238, 2021.
  3. L. Ma, T. Ma, R. Liu, X. Fan, Z. Luo, "Toward Fast, Flexible, and Robust Low-light Image Enhancement." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 5637-5646, 2022.
  4. J. Redmon, S. Divvala, R. Girshick, A. Farhadi, "You Only Look Once: Unified, Real-time Object Detection." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 779-788, 2016.
  5. J. Deng, J. Guo, E. Ververas, I. Kotsia, S. Zafeiriou, "Retinaface: Single-shot Multi-level Face Localisation in the Wild." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 5203-5212, 2020.
  6. K. Zhang, Z. Zhang, Z. Li, Y. Qiao, "Joint Face Detection and Alignment Using Multitask Cascaded Convolutional Networks." IEEE Signal Processing Letters, Vol. 23, No. 10, pp. 1499- 1503, 2016.
  7. W. Liu, Y. Wen, Z. Yu, M. Yang, "Large-margin Softmax Loss for Convolutional Neural Networks." arXiv preprint arXiv:1612.02295, 2016.
  8. E. Hoffer, N. Ailon, "Deep Metric Learning Using Triplet Network." In: Similarity-Based Pattern Recognition: Third International Workshop, SIMBAD 2015, Copenhagen, Denmark, October 12-14, 2015. Proceedings 3. Springer International Publishing, pp. 84-92, 2015.
  9. J. Deng, J. Guo, N. Xue, S. Zafeiriou, "Arcface: Additive Angular Margin Loss for Deep Face Recognition." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 4690-4699, 2019.
  10. C. Wei, W. Wang, W. Yang, J. Liu, "Deep Retinex Decomposition for Low-light Enhancement." arXiv preprint arXiv:1808.04560, 2018.
  11. Y. Zhang, J. Zhang, X. Guo, "Kindling the Darkness: A Practical Low-light Image Enhancer." Proceedings of the 27th ACM International Conference on Multimedia, pp. 1635-1643, 2019.
  12. E. H. Land, J. J. McCann, "Lightness and Retinex Theory." Journal of the Optical Society of America, Vol. 61, No. 1, pp. 1-11, 1971.
  13. G. P. Nam, H. S. Choi, J. H. Cho, I. J. Kim, "Large-Scale Korean Face Database Construction," Proceedings of the IEIE Conference, pp. 62-63, 2017 (in Korean).
  14. Y. Guo, L. Zhang, Y. Hu, X. He, J. Gao, "MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition." Computer Vision-ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part III 14. Springer International Publishing, pp. 87-102, 2016.
  15. S. Chen, Y. Liu, X. Gao, Z. Han, "Mobilefacenets: Efficient cnns for Accurate Real-time Face Verification on Mobile Devices." Biometric Recognition: 13th Chinese Conference, CCBR 2018, Urumqi, China, August 11-12, 2018, Proceedings 13. Springer International Publishing, pp. 428-438, 2018.
  16. https://en.wikipedia.org/wiki/Receiver_operating_characteristic#Curves_in_ROC_space