After detecting the face, the eye regions are to be extracted. Eyes are the most important feature of human face. Since, eyes lie in the upper half of the face, the lower part of face is removed in order to reduce search area. There are many applications of the robust eye states extraction [8]. For example, the eye states provide important information for recognizing facial expression and human-computer interface systems. In this paper, a method is presented to detect the eye states in nearly frontal-view color face image sequence by using the Hough transform. The eye is the most significant and important feature in a human face, as extraction of the eyes are often easier as compared to other facial features. Eye detection is done by using various methods [9]. The method which is used in this paper for detection of eye is the circular Hough transform. After detecting the face from input video frame next step is to extract the eyes. In order to reduce the search area a dummy image is created, which is used as a mask over the image. And then fill this dummy image by using imfill command which fills the holes in the dummy image. In order to get the eye region from the search area multiply the dummy image with that of the face detected image, whose output becomes the area which contains the eye part. Result of this are as follows