This paper proposes the concept of Machine Learning to implement an eye blink detection system. It is used to first track the eyes on the patient and then detect its movements. Machine learning has experienced a boost in acceptance among many fields including the medical field. People suffering from speech impairment find it difficult to communicate their needs to the outside world. People with severe disabilities are confined in a state in which communication is virtually impossible, being reduced to communicating with their eyes or using sophisticated systems that translate thoughts into words. The Eye Trackers are suitable systems for those people but the main disadvantage is the cost. More affordable devices are capable of detecting voluntary blinks and translating them into a binary signal that allows the selection. The method of voluntary blinking, the use of long or double blinks had no statistical influence on accuracy, excluding EOG, and the time taken to perform double blinks was shorter, resulting in a potentially much faster interface. Machine learning creates a platform to be precise in the measurement of any parameters using various algorithms. In this paper, we propose to apply Haar cascade and shape predictor algorithms to map the eyes of the patient and detect various blink patterns. The preferred technology and blinking methods were Video-Oculography (VOG) and long blinks Implementation of this paper successfully bridges the communication gap between the outside world and the paralyzed/disabled patients.

Open CV, eye aspect ratio, Ada boost classifier, face detection, EOG, VOG

[1]. Joshua D. Fischer, Dawie J. van den Heever, ?Portable Video-Oculography Device for Implementation in Sideline Concussion Assessments: A Prototype?, IEEE conference, 2016.
[2]. Sudhir Rao Rupanagudi, Vikas N S, Vivek C Bharadwaj, Manju; Dhruva N; Sowmya K. S, ?Novel methodology for blink recognition using video oculography for communicating?, IEEE Conference Publications, ICAEE, 2014.
[3]. Masaru Kiyama, Hitoshi Iyatomi and Koichi Ogawa, ?Development of robust videooculography system for non-invasive automatic nerve quantification?, IEEE Conference Publications, IEEE-EMBS Conference on Biomedical Engineering and Sciences, 2012.
[4]. S. M. H. Jansen, H. Kingma, R. L. M. Peeters and R. L. Westra, ?A torsional eye movement calculation algorithm for low contrast images in video-oculography ?, IEEE
[5]. Akshat Gambhir, K.S Boppaiah, M Shruthi Subbaiah, Pooja M and KiranP, RNSIT, ?Video Oculographic System using Real-Time Video Processing?, International Journal of Computer Applications (0975 ?8887) Volume 119 ?No.22, June 2015.
[6]. Karthik K.P and Basavaraju S, Department of Electronics & Communication Engineering, Sapthagiri College of Engineering, Bangalore, ?Design and Implementation of Unique Video Oculographic Algorithm Using Real Time Video Processing on FPGA?, IJSRD -International Journal for Scientific Research & Development| Vol. 3, Issue 04, 2015 | ISSN (online): 2321-0613.
[7]. C. D. N. Ayudhya and T. Srinark, ??A method for real-time eye blink detection and its application,?? in Proc. 6th Int. Joint Conf. Comput. Sci.Softw. Eng. (JCSSE), pp. 1?6, 2009.
[8]. P. Biswas and P. Langdon, ??A new interaction technique involving eye gaze tracker and scanning system,?? in Proc. Conf. Eye Tracking South Africa, pp. 67?70, 2013.
[9]. M. Chau and M. Betke, ??Real time eye tracking and blink detection with USB cameras,?? Dept. Comput. Sci., Boston Univ., Boston, MA, USA,Tech. Rep. 2005-12, 2005.
[10]. E. Dalmaijer, ??Is the low-cost EyeTribe eye tracker any good forresearch??? PeerJ PrePrints, vol. 2, , Art. no. e585v1, 2014.
[11]. A. Dementyev and C. Holz, ??Dualblink: A wearable device to continuously detect, track, and actuate blinking for alleviating dry eyesand computer vision syndrome,?? Proc. ACM Interact. Mobile Wearable Ubiquitous Technol., vol. 1, no. 1, pp. 1:1?1:19, Mar. 2017.
[12]. M. Divjak and H. Bischof, ??Eye blink based fatigue detection for prevention of computer vision syndrome,?? in Proc. MVA, pp. 350?353, 2009.
[13]. T. Drutarovsky and A. Fogelton, ??Eye blink detection using varianceof motion vectors,?? in Proc. Eur. Conf. Comput. Vis. Springer, pp. 436?448, 2014.
[14]. K. Wang and Q. Ji, ??Real time eye gaze tracking with Kinect,??in Proc. 23rd Int. Conf. Pattern Recognit. (ICPR), pp. 2752?2757, Dec. 2016.
[15]. R. Valenti and T. Gevers, ??Accurate eye center location and tracking using isophote curvature,?? in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), pp. 1?8, Jun. 2008.
[16]. X. Zheng, X. Li, J. Liu, W. Chen, and Y. Hao, ??A portable wirelesseye movement-controlled human-computer interface for the disabled,?? in Proc. Int. Conf. Complex Med. Eng. (ICME), Tempe, AZ, USA, pp. 1?5, Apr. 2009.
[17]. T. Soukupov?, ??Real-time eye blink detection using facial landmarks,??in Proc. 21st Comput. Vis. Winter Workshop Luka Cehovin, R. Mandeljc Rimske Toplice, Feb. 20