Pupil Detection in Facial Images with Using Bag of Pixels

Abstract:

To detect and track eye images, distinctive features of user eye are used. Generally, an eye-tracking and detection system can be divided into four steps: Face detection, eye region detection, pupil detection and eye tracking. To find the position of pupil, first, face region must be separated from the rest of the image using bag of pixels, this will cause the images background to be non effective in our next steps. We used from horizontal projection, to separate a region containing eyes and eyebrow. This will result in decreasing the computational complexity and ignoring some factors such as bread. Finally, in proposed method points with the highest values of are selected as the eye candidate's. The eye region is well detected among these points. Color entropy in the eye region is used to eliminate the irrelevant candidates. With a pixel of the iris or pupil can be achieved center of pupil. To find the center of pupil can be used line intersection method in the next step, we perform eye tracking. The proposed method achieve a correct eye detection rate of 97.3% on testing set that gathered from different images of face data. Moreover, in the case of glasses the performance is still acceptable.

• Periodical: Advanced Materials Research (Volumes 468-471)
• Pages: 2941-2948
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.468-471.2941
• Online since: February 2012
• Authors: Mohammad Ali Azimi Sotudeh, Hasan Ziafat, Said Ghafari
• Keywords: Bag of Pixels, Cam Shift, Eye Tracking, Pupil Detection

References

[1] D. Hansen Q.Ji. (2010). In the Eye of the Beholder: A Survey of Models for Eyes and Gaze. Pattern Analysis and Machine Intelligence, IEEE Transactions

DOI: 10.1109/tpami.2009.30

[2] F.Ehsan and John. , (2005). Rotaion invarient facial feature detection using gabor wavelet and entropy.LNCS 3656, pp.1040-1047.

[3] M.Ramezanpour, M.Rahmati, M.Assari. , (2010). Image Processing Method for Real-time Eye Detection. International Conference on Image Processing, Computer Vision, & Pattern Recognition (IPCV'11), USA.

[4] J.Song, Z.Chi. (2006). A robust eye detection method using combined binary and intensity information. Pattern recognition, Vol 39.

DOI: 10.1016/j.patcog.2005.11.015

[5] E. Rosten and T. Drummond. (2005). Fusing Points and Lines for High Performance Tracking. Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 2, pp.1508-1515.

DOI: 10.1109/iccv.2005.104

[6] T. Jebara. (2003). Convex invariance learning. In Articial Intelligence and Statistics 9.

[7] Yi Guo and Junbin GAO. (2006). Manifolds of Bag of Pixels: A Better Representation for Image Recognition?. IEEE International Conference on Systems, Man, and Cybernetics October 8-11, Taipei, Taiwan.

DOI: 10.1109/icsmc.2006.384691

[8] M.A Azimi Kashani, M.Ramezanpour.(2011). Face Image Retrieval Using Genetic Algorithm and Bags of Pixels. In IEEE Eighth International Conference on Information Technology: New Generations, USA.

DOI: 10.1109/itng.2011.91

[9] M.Ramezanpour, M.A Azimi Kashani, M.Rahmati. (2011). Eye detection and tracking in image with complex background. In IEEE 3rd International Conference on Electronics Computer Technology (ICECT 2011), Kanyakumari, India.

DOI: 10.1109/icectech.2011.5942050

[10] A.Azimi Kashani, M.Mollaie, M.Ramezanpour. (2011).An algorithm for Eye detection and tracking in face images, International Conference on Image Processing, Computer Vision, & Pattern Recognition (IPCV'11), USA.