Research on Robust Face Recognition against Lighting Variation using CNN

CNN을 적용한 조명변화에 강인한 얼굴인식 연구

  • Received : 2017.02.02
  • Accepted : 2017.04.24
  • Published : 2017.04.30


Face recognition technology has been studied for decades and is being used in various areas such as security, entertainment, and mobile services. The main problem with face recognition technology is that the recognition rate is significantly reduced depending on the environmental factors such as brightness, illumination angle, and image rotation. Therefore, in this paper, we propose a robust face recognition against lighting variation using CNN which has been recently re-evaluated with the development of computer hardware and algorithms capable of processing a large amount of computation. For performance verification, PCA, LBP, and DCT algorithms were compared with the conventional face recognition algorithms. The recognition was improved by 9.82%, 11.6%, and 4.54%, respectively. Also, the recognition improvement of 5.24% was recorded in the comparison of the face recognition research result using the existing neural network, and the final recognition rate was 99.25%.


Supported by : 순천대학교


  1. L. Yann, B. Yoshua, and G. Hinton, "Deep leaning," Nature, vol. 521, no. 7553, 2015, pp. 436-444.
  2. A. Krizhevsky, I. Sutskever, and G. E. Hinton, "Imagenet classification with deep convolutional neural networks," Advances in Neural Information Processing Systems, vol. 21, no. 1, 2012, pp. 1097-1105.
  3. D. Gowe, "Distinctive image features from scale-invariant key points," Int. J. of Computer Vision. vol. 60, no. 2, 2004, pp. 91-110.
  4. H. Bay, T. Tuytelaars, and L. Vanl, "Surf: Speeded up robust features," European Conf. on Computer Vision, Heidelberg, Germany, May 2006, pp. 404-417.
  5. Y. Jia, E. Shelhamer, J. Donahue, S. Karayev, J. Long, R. Girshick, S. Guadarrama, and T. Darrell, "Caffe: An open source convolutional architecture for fast feature embedding," In Proc. the 22nd ACM Int Con. on Multimedia, Orlando, USA, Nov. 2014, pp. 675-678.
  6. K. Simonyan and A. Zisserman, "Very deep convolutional networks for large-scale image recognition," Proc. of Int. Conf. on Learning Representations, San Diego, USA, Jun, 2015 pp. 1-14,
  7. A. Georghiades, P. Nelhumeur, and D. Kriegman, "From few to many: Illumination cone models for face recognition under variable lighting and pose," IEEE Trans Pattern Analysis and Machine Intelligence, vol. 23, no. 6, 2001, pp. 643-660.
  8. L. Luo, M. Swamy, and E. Plotkin, "A modified PCA algorithm for face recognition," In Canadian Conf. on Electrical and Computer Engineering, Quebec, Canada, vol. 1, May 2003, pp. 57-60.
  9. J. Lee, "A Study on Face Recognition System Using LDA and SVM," J. of the Korea Institute of Electronic Communication Sciences, vol. 10, no. 11, 2015, pp. 1307-1314.
  10. A. Petpon and S. Srisuk, "Face Recognition with Local Line Binary Pattern," In Int. Conf. on Image and Graphics, Shanxi, China, vol. 9, Mar. 2009, pp. 533-539.
  11. T. Ahonen, A. Hadid, and M. Pietikainen, "Face description with local binary patterns: application to face recognition," IEEE Trans Pattern Analysis and Machine Intelligence, vol. 28, no. 12, 2006, pp. 2037-2041.
  12. X. Xiaohua, Z. Wei-shi, L. Jianhuang, and Y. Pong C, "Face illumination normalization on large and small scale features," In IEEE Conf. on Computer Vision and Pattern Recognition, Anchorage, USA, 2008, pp. 1-5.
  13. H. Lien and A. Jain. "Face modeling for recognition," In Proc. of the Int. Conf. on Image Processing, Thessaloniki, Greece, vol. 2, Oct. 2001, pp. 693-696.
  14. Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, "Gradient-based learning applied to document recognition," In Proc the IEEE, vol. 86, no.11, 1998, pp. 2278-2324.
  15. Y. Lee, "A Comparison and Analysis of Deep Learning Framework," J. of the Korea Institute of Electronic Communication Sciences, vol. 12, no. 1, 2017, pp. 115-122.