Iris super-resolution using CNNs: is photo-realism important to iris recognition?
- Author(s): Eduardo Ribeiro 1, 2 ; Andreas Uhl 1 ; Fernando Alonso-Fernandez 3
-
-
View affiliations
-
Affiliations:
1:
Department of Computer Sciences , University of Salzburg , Jakob Haringer Strasse 2 5020 , Salzburg , Austria ;
2: Department of Computer Sciences , Federal University of Tocantins , 109 Norte , Av. NS 15 , ALC NO 14 , Palmas , Brazil ;
3: IS-Lab/CAISR , Halmstad University , Box 823 , Halmstad SE 301-18 , Sweden
-
Affiliations:
1:
Department of Computer Sciences , University of Salzburg , Jakob Haringer Strasse 2 5020 , Salzburg , Austria ;
- Source:
Volume 8, Issue 1,
January
2019,
p.
69 – 78
DOI: 10.1049/iet-bmt.2018.5146 , Print ISSN 2047-4938, Online ISSN 2047-4946
The use of low-resolution images adopting more relaxed acquisition conditions such as mobile phones and surveillance videos is becoming increasingly common in iris recognition nowadays. Concurrently, a great variety of single image super-resolution techniques are emerging, especially with the use of convolutional neural networks (CNNs). The main objective of these methods is to try to recover finer texture details generating more photo-realistic images based on the optimisation of an objective function depending basically on the CNN architecture and training approach. In this work, the authors explore single image super-resolution using CNNs for iris recognition. For this, they test different CNN architectures and use different training databases, validating their approach on a database of 1.872 near infrared iris images and on a mobile phone image database. They also use quality assessment, visual results and recognition experiments to verify if the photo-realism provided by the CNNs which have already proven to be effective for natural images can reflect in a better recognition rate for iris recognition. The results show that using deeper architectures trained with texture databases that provide a balance between edge preservation and the smoothness of the method can lead to good results in the iris recognition process.
Inspec keywords: image texture; iris recognition; video signal processing; image resolution; edge detection; visual databases; feedforward neural nets; realistic images; feature extraction
Other keywords: low-resolution images; photo-realism; near infrared iris images; iris super-resolution; mobile phone image database; texture databases; quality assessment; iris recognition process; training databases; edge preservation; single image super-resolution techniques; natural images; convolutional neural networks; CNN architectures; texture recovery; photorealistic image generation; objective function optimisation; surveillance videos
Subjects: Image recognition; Computer vision and image processing techniques; Neural computing techniques; Video signal processing
References
-
-
1)
-
37. Fei-Fei, L., Fergus, R., Perona, P.: ‘Learning generative visual models from few training examples: an incremental Bayesian approach tested on 101 object categories’, Comput. Vis. Image Underst., 2007, 106, (1), pp. 59–70.
-
-
2)
-
44. Zhang, L., Zhang, L., Mou, X., et al: ‘FSIM: a feature similarity index for image quality assessment’, IEEE Trans. Image Process., 2011, 20, (8), pp. 2378–2386.
-
-
3)
-
9. Yang, C.-Y., Ma, C., Yang, M.-H.: ‘Single-image super-resolution: a benchmark’ (Springer International Publishing, Cham, 2014), pp. 372–386.
-
-
4)
-
32. Hofbauer, H., Alonso-Fernandez, F., Wild, P., et al: ‘A ground truth for iris segmentation’. 2014 22nd Int. Conf. on Pattern Recognition, Stockholm, Sweden, August 2014, pp. 527–532.
-
-
5)
-
1. Baker, S., Kanade, T.: ‘Limits on super-resolution and how to break them’, IEEE Trans. Pattern Anal. Mach. Intell., 2002, 24, (9), pp. 1167–1183.
-
-
6)
-
8. Nasrollahi, K., Moeslund, T.B.: ‘Super-resolution: a comprehensive survey’, Mach. Vis. Appl., 2014, 25, (6), pp. 1423–1468.
-
-
7)
-
36. Dana, K., Van Ginneken, B., Nayar, S., et al: ‘Reflectance and texture of real-world surfaces’, ACM Trans. Graph., 1999, 18, (1), pp. 1–34.
-
-
8)
-
12. Mjolsness, E.: ‘Neural networks, pattern recognition, and fingerprint hallucination’. PhD thesis (dissertation), California Institute of Technology, 1986.
-
-
9)
-
45. Lowe, D.G.: ‘Distinctive image features from scale-invariant keypoints’, Int. J. Comput. Vis., 2004, 60, (2), pp. 91–110.
-
-
10)
-
46. Alonso-Fernandez, F., Tome-Gonzalez, P., Ruiz-Albacete, V., et al: ‘Iris recognition based on SIFT features’. 2009 First IEEE Int. Conf. on Biometrics, Identity and Security (BIdS), Tampa, Florida, September 2009.
-
-
11)
-
26. Ribeiro, E., Uhl, A.: ‘Exploring texture transfer learning via convolutional neural networks for iris super resolution’. LNI, GI/IEEE Proc. 2017 Int. Conf. of the Biometrics Special Interest Group (BIOSIG'17), Darmstadt, Germany, 2017.
-
-
12)
-
27. Ribeiro, E., Uhl, A., Alonso-Fernandez, F., et al: ‘Exploring deep learning image super-resolution for iris recognition’. 2017 Proc. 25th European Signal Processing Conf. (EUSIPCO 2017), Kos Island, Greece, 28 August–2 September 2017.
-
-
13)
-
33. Burghouts, G., Geusebroek, J.: ‘Material-specific adaptation of color invariant features’, Pattern Recognit. Lett., 2009, 30, (3), pp. 306–313.
-
-
14)
-
3. Nguyen, K., Fookes, C., Sridharan, S., et al: ‘Feature-domain super-resolution for iris recognition’, Comput. Vis. Image Underst., 2013, 117, (10), pp. 1526–1535.
-
-
15)
-
18. Nguyen, K., Fookes, C., Sridharan, S., et al: ‘Quality-driven super-resolution for less constrained iris recognition at a distance and on the move’, IEEE Trans. Inf. Forensics Sec., 2011, 6, (4), pp. 1248–1258.
-
-
16)
-
24. Shin, K.Y., Park, K.R., Kang, B.J., et al: ‘Super-resolution method based on multiple multi-layer perceptrons for iris recognition’. Proc. Fourth Int. Conf. on Ubiquitous Information Technologies Applications, Fukuoka, Japan, December 2009, pp. 1–5.
-
-
17)
-
4. Kalka, N.D., Zuo, J., Schmid, N.A., et al: ‘Estimating and fusing quality factors for iris biometric images’, IEEE Trans. Syst. Man Cybern. A, Syst. Humans, 2010, 40, (3), pp. 509–524.
-
-
18)
-
42. Alonso-Fernandez, F., Farrugia, R.A., Bigun, J.: ‘Learning-based local-patch resolution reconstruction of iris smartphone images’. IEEE/IAPR Int. Joint Conf. on Biometrics, IJCB, Denver, Colorado, October 2017.
-
-
19)
-
7. Alonso-Fernandez, F., Farrugia, R.A., Bigun, J.: ‘Iris super-resolution using iterative neighbor embedding’. 2017 IEEE Conf. on Computer Vision and Pattern Recognition Workshops (CVPRW), Honolulu, Hawaii, USA, July 2017, pp. 655–663.
-
-
20)
-
10. Allebach, J., Wong, P.W.: ‘Edge-directed interpolation’. Proc. Third IEEE Int. Conf. on Image Processing, Lausanne, Switzerland, Switzerland, 1996, vol. 3, pp. 707–710.
-
-
21)
-
22. Alonso-Fernandez, F., Farrugia, R.A., Bigun, J.: ‘Improving very low-resolution iris identification via super-resolution reconstruction of local patches’. 2017 Int. Conf. of the Biometrics Special Interest Group (BIOSIG), Darmstadt, Germany, September 2017, pp. 1–6.
-
-
22)
-
5. Jonathon Phillips, P., Patrick Flynn, J., Ross Beveridge, J., et al: ‘Overview of the multiple biometrics grand challenge’ (Springer, Berlin, Heidelberg, 2009), pp. 705–714.
-
-
23)
-
23. Nguyen, K., Sridharan, S., Denman, S., et al: ‘Feature-domain super-resolution framework for Gabor-based face and iris recognition’. 2012 IEEE Conf. on Computer Vision and Pattern Recognition, Providence, RI, USA, 16–21 June 2012, pp. 2642–2649.
-
-
24)
-
16. Kim, J., Lee, J.K., Lee, K.M.: ‘Accurate image super-resolution using very deep convolutional networks’. Computing Research Repository (CoRR), Ithaca, NY, 2015, abs/1511.04587.
-
-
25)
-
19. Deshpande, A., Patavardhan, P.P.: ‘Super resolution and recognition of long range captured multi-frame iris images’, IET Biometrics, 2017, 6, (5), pp. 360–368.
-
-
26)
-
20. Cui, J., Wang, Y., Huang, J.Z., et al: ‘An iris image synthesis method based on PCA and super-resolution’. 2004 Proc. 17th Int. Conf. on Pattern Recognition (ICPR 2004), Cambridge, UK, August 2004, vol. 4, pp. 471–474.
-
-
27)
-
31. Goodfellow, I., Pouget-Abadie, J., Mirza, M., et al: ‘Generative adversarial nets’, in Ghahramani, Z., Welling, M., Cortes, C., Lawrence, N.D., Weinberger, K.Q. (eds.): ‘Advances in neural information processing systems’ (Curran Associates Inc., 2014), vol. 27, pp. 2672–2680.
-
-
28)
-
11. Li, X., Orchard, M.T.: ‘New edge-directed interpolation’, IEEE Trans. Image Process., 2001, 10, (10), pp. 1521–1527.
-
-
29)
-
6. Nguyen, K., Sridharan, S., Denman, S., et al: ‘Feature-domain superresolution framework for Gabor-based face and iris recognition’. 2012 IEEE Conf. on Computer Vision and Pattern Recognition, Providence, Rhode Island, June 2012, pp. 2642–2649.
-
-
30)
-
41. Wang, N., Tao, D., Gao, X., et al: ‘A comprehensive survey to face hallucination’, Int. J. Comput. Vis., 2014, 106, (1), pp. 9–30.
-
-
31)
-
21. Fahmy, G.: ‘Super-resolution construction of iris images from a visual low resolution face video’. 2007 9th Int. Symp. on Signal Processing and its Applications, Sharjah, United Arab Emirates, February 2007, pp. 1–4.
-
-
32)
-
25. Zhang, Q., Li, H., He, Z., et al: ‘Image super-resolution for mobile iris recognition’ (Springer International Publishing, Cham, Switzerland, 2016), pp. 399–406.
-
-
33)
-
13. Wang, Z., Liu, D., Yang, J., et al: ‘Deeply improved sparse coding for image super-resolution’. Computing Research Repository (CoRR), Santiago, Chile, 2015, abs/1507.08905.
-
-
34)
-
30. LeCun, Y., Bottou, L., Bengio, Y., et al: ‘Gradient-based learning applied to document recognition’, ‘Intelligent signal processing’ (IEEE Press, Piscataway, NJ, 2001), pp. 2278–2324.
-
-
35)
-
14. Dong, C., Loy, C.C., He, K., et al: ‘Image super-resolution using deep convolutional networks’. Computing Research Repository (CoRR), Ithaca, NY, 2015, abs/1501.00092.
-
-
36)
-
34. Cimpoi, M., Maji, S., Kokkinos, I., et al: ‘Describing textures in the wild’. Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), Columbus, 2014.
-
-
37)
-
35. Sharana, L., Rosenholtz, R., Adelson, E.: ‘Material perception: what can you see in a brief glance?’, J. Vis., 2009, 9, p. 784.
-
-
38)
-
28. Vedaldi, A., Lenc, K.: ‘MatConvNet – convolutional neural networks for MATLAB’. Computing Research Repository (CoRR), Ithaca, NY, 2014, abs/1412.4564.
-
-
39)
-
40. Raja, K.B., Raghavendra, R., Vemuri, V.K., et al: ‘Smartphone based visible iris recognition using deep sparse filtering’, Pattern Recognit. Lett., 2015, 57, (Supplement C), pp. 33–42, Mobile Iris CHallenge Evaluation part I (MICHE I).
-
-
40)
-
15. Ledig, C., Theis, L., Huszar, F., et al: ‘Photo-realistic single image super-resolution using a generative adversarial network’. Computing Research Repository (CoRR), Ithaca, NY, 2016, abs/1609.04802.
-
-
41)
-
2. Park, S.C., Park, M.K., Kang, M.G.: ‘Super-resolution image reconstruction: a technical overview’, IEEE Signal Process. Mag., 2003, 20, (3), pp. 21–36.
-
-
42)
-
29. He, K., Zhang, X., Ren, S., et al: ‘Deep residual learning for image recognition’. Computing Research Repository (CoRR), Ithaca, NY, 2015, abs/1512.03385.
-
-
43)
-
39. Rathgeb, C., Uhl, A., Wild, P., et al: ‘Design decisions for an iris recognition sdk’, in Bowyer, K., Burge, M.J. (eds.): ‘Handbook of iris recognition’, Advances in Computer Vision and Pattern Recognition (Springer-Verlag, London, UK, 2016, 2nd edn.), pp. XXI.
-
-
44)
-
43. Chen, H.Y., Chien, S.Y.: ‘Eigen-patch: position-patch based face hallucination using eigen transformation’. 2014 IEEE Int. Conf. on Multimedia and Expo (ICME), Bandung, Indonesia, July 2014, pp. 1–6.
-
-
45)
-
17. Simonyan, K., Zisserman, A.: ‘Very deep convolutional networks for large-scale image recognition’. Computing Research Repository (CoRR), Ithaca, NY, 2014, abs/1409.1556.
-
-
46)
-
38. Ribeiro, E., Barcelos, C., Batista, M.: ‘Image characterization via multilayer neural networks’. 2008 20th IEEE Int. Conf. on Tools with Artificial Intelligence, Dayton, Ohio, November 2008, vol. 1, pp. 325–332.
-
-
1)