The number of casualties from road accidents keeps arising each year. During long distance journeys , drivers usually are sleep deprived as they do not take enough breaks and there is a high risk of becoming drowsy and causing accidents. Drowsy Driver Detection System is developed, using a machine learning and image processing concepts. Proposed system is ready to use for car because of compact hardware which consumes less power and provides much faster processing speed and also can be operated in low light conditions. This system detects the eye landmarks from camera frames for monitoring the drowsiness of driver. Inspite of having several methods for measuring the drowsiness, this approach is completely non-intrusive and does not affect the driver in any way, thus giving the exact condition of the driver. For detection of drowsiness the method of per closure value of eye is considered. When the closure of eye exceeds a certain amount then the driver is identified to be sleepy. For implementing the system several OpenCv libraries are used including Haar-cascade. The entire system is implemented using Raspberry-Pi.