Controlling devices using facial expression identification and recognition for assisting physically disabled

摘要

In the present scenario there is a need to do something for the society mainly for differently abled people. In this paper we implement the control of devices through tongue movements to aid the differently abled people. A pat of four tongue movements namely left, right, top, and down is used to control pre-defined devices. The pre-dominant camcorder acquires the video of the subject. A small luminous black sticker is fixed on subject's tongue. The Viola-Jones detection algorithm is used to detect the subject's tongue from the real time video. In order to have a better quality of the image the histogram equalization and filtering is done. Afterwards the analysis of Co-ordinate System (CSA) is used to detect the movement of the black marker sticker placed on the tongue. The pre-defined tongue movement is exclusively represented with the help of 8-bit binary number. The serial port communicates to the microcontroller and the devices are controlled accordingly.