<![CDATA[Disability is defined as a disability or inability to do something. There are various kinds of people with disabilities, one of whom is a person with a physical disability. In controlling a device with physical disabilities in the hands and feet it will have an impact on their difficulty in mobilizing so that they still need help from other people or other devices. These problems can be solved by using a camera-based control control system instead of the joystick or remote as a control tool. This study was designed to detect eye position based on head movement through image processing using cameras and facial landmarks as a method of detection. Processing on this system uses Raspberry Pi 3B + and trigonometric analysis. The output implementation on this system uses a servo motor. Tests on this system by testing 8 subjects using different time and distance from the position of face objects. The results of this study are the best time to detect the eye at any position of head movement in the morning and afternoon. The percentage of the best head movement is straight, right angled, left and left angled with a 100% accuracy. While on the right tilt movement get a percentage of 98.96%. System computing time is determined by the distance of the subject in front of the camera. The farther the subject from the front of the camera, the computing time of the system will be smaller. Conversely, if the subject is getting closer, the time for computation is getting bigger. For accuracy System integration with hardware is based on the classification output on the system. If based on the original movement, the accuracy is the same as the accuracy in detecting head movement based on the eye position.]]>
Copyrights © 2019
