Eye blinking detection to perform selection for an eye tracking system used in assistive technology

The paper presents the analysis of methods used to implement eye blinking selection in an eye tracking system used in assistive technology for patients with neuro-motor disabilities. The system uses for selection the key words technology or ideograms presented on the user's screen. The method used for blinking detection is based on image segmentation using an adaptive local threshold determined using the integral sum image or Bradley method. Results obtained present that the method implemented for blinking detection can be used efficiently in a real time eye tracking system for assistive technology applications.