An appearance cum feature based iris center gaze detection model viz. Iris Center Based Gaze Estimation (ICGE) is proposed to detect the direction of gaze quadrants to overcome certain limitations like dependency on light sources, multiple glint formation, no formation of glint etc, observed in glint based gaze quadrant detection models. The model works on the adaptive thresholding technique for the detection of the iris center coordinates using more than two hundred images from an indigenous database of different subjects on a five quadrants map screen. The model works with the Circular Hough Transform (CHT) for localising circles in the eye images and then center coordinates on the iris edge for further detection of gaze quadrants. Gaze directions of five different positions of iris are estimated on a mapped screen within the eye region. The model generates almost ninety percent accurate results for correct iris and gaze quadrant detection. The distinguishing features of the low cost, non intrusive proposed model include non IR and affordable ubiquitous hardware designing, large subject-camera distance and screen dimensions, no glint dependency etc. The proposed model also shows significantly better results in the lower periphery corners of the quadrant map. The proposed model may be more suitable for interactive applications for healthy users who cannot use head and hands freely while doing other tasks or disabled users who have no movement in their hands and head etc.

Iris Center Based Gaze Estimation (ICGE) model, adaptive thresholding, iris center, gaze quadrant detection

[1] D.W. Hansen and Q. Ji, “In the eye of the beholder: A survey of models for eyes and gaze”, IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 32, No. 3, pp. 478-500, 2010.
[2] C. Djeraba, A. Labelak, and Y. Benabbas, “Estimation of visual gaze, In Multi-Modal User Interactions in Controlled Environments”, chap. 4, pp. 99-141, Springer Science + Business Media, New York, 2010.
[3] M. Ohtani and Y. Ebisawa, “Eye gaze detection based on the pupil detection technique using two light sources and the image difference method”, IEEE-EMBC and CMBEC, Theme 7: Instrumentation, pp. 1623-1624,1995.
[4] D. Kaur et al., “Various image segmentation techniques: A review”, International Journal of Computer Science and Mobile Computing, Vol. 3, No. 5, pp. 809-814, 2014.
[5] L. Świrski, A. Bulling, and N. Dodgson, “Robust real-time pupil tracking in highly off-axis images”, In the Proceedings of the Symposium on Eye Tracking Research and Applications, pp. 173-176, 2012.
[6] J. Sigut and S. Sidha, “Iris center corneal reflection method for gaze tracking using visible light”, IEEE Trans. on Biomed. Eng., Vol 58, Issue 2, pp. 411-419, 2010, [doi: 10.1109/ TBME.2010.2087330].
[7] D. Kimmel, D. Mammo, and W. Newsome, “Tracking the eye non-invasively: simultaneous comparison of the scleral search coil and optical tracking techniques in the macaque monkey”, Frontiers in Behavioural Neuroscience, Vol. 6, No. 49, pp. 312-331, 2012.
[8] P. D. Prasad, “Iris recognition”, Department of Computer Science & Engineering, National Institute of Technology, Thesis for Bachelor of Technology, 2005.
[9] A. Sharma and P. Abrol, “Direction estimation model for gaze controlled systems”, Journal of Eye Movement Research, Vol. 9, Issue 6, No. 5, pp. 1-12, 2016, [doi: 10.16/910//jemr.9.6.5].
[10] J. W. Ryan et al., “Adapting starburst for elliptical iris segmentation”, In the Proceedings of IEEE Second International Conference on Biometrics: Theory, Applications and Systems, 2008.
[11] S. Shah and A. Ross, “Iris segmentation using geodesic active contours,” IEEE Trans. on information forensic and security, Vol 4, Issue 4, 2009.
[12] "Adaptive thresholding," (Accessed on 15 Dec 2016). [http://www.123HelpMe.com/view.asp?Id= 50915]
[13] A. Sharma and P. Abrol, “Comparative analysis of edge detection operators for better glint detection”, In the Proceedings of 9th INDIACOM in 2nd International Conference on Computing for Sustainable Global Development, IEEExplore digital library. ISBN 978-9-3805-4415-1, pp. 973-977, 2015.
[14] T. Moravcik, “An approach to iris and pupil detection in eye image”, XII International PhD Workshop OWD, pp. 239-242, 2010.
[15] W.M Khairosfaizal and.A.J. Noraini, “Eyes detection in facial images using Circular Hough Transform”, In the Proceedings of 5th International Colloquium on Signal Processing & Its Applications, CSPA, pp. 238-242, 2009.

[16] N. Avazpour and M. Emadi, “Iris recognition methods: A Review” International Journal of Computer Science and Information Security (IJCSIS), Vol. 14, No. 11, pp. 527-534, 2016.
[17] S. M. Zadeh and A. Harimi, “Iris localization by means of adaptive thresholding and Circular Hough Transform ”, Journal of Artificial Intelligence and Data Mining, Vol. 5, No. 1, pp. 21-28, 2017, [doi: 10.22044/jadm.2016.731]
[18] M. Abdullah, S. Dlay, W. Woo and J. Chambers, “Robust iris segmentation method based on a new active contour force with a non-ideal normalization”, IEEE Transactions on Systems, Man, and Cybernetics: Systems, Vol. 99, pp. 1-14, 2016, [doi: 10.1109/ TSMC.2016.2562500].
[19] Y. Mingxin et al., “An easy iris center detection method for eye gaze tracking system,” Journal of Eye Movement Research, Vol. 8, issue 3, No. 5, pp. 1-20, 2015, [doi: 10.16910/jemr.8.3.5].
[20] A. Sharma and P. Abrol, “Evaluating Interactivity with respect to Distance and Orientation Variables of GDE Model,” In the Proceedings of the Springer International Congress on Information and Communication Technology (Advances in Intelligent Systems and Computing), 438, Springer Singapore, pp. 177-187, 2015, [doi: 10.1007/978-981-10-0767-5_20].
[21] D. Bradley and G. Roth, “Adaptive thresholding using the integral image,” Carleton University, Canada, Journal of Graphics Tools, NRC 48816, Vol. 12, issue 2, pp. 13-21, 2007.