Recognizing Driver Somnolence using Computer Vision
Anitha K1, Nischitha V N2, Prakruthi SP3, Sindhu D K4
1Anitha K, Assistant Professor, Department of Computer Science, Raja Rajeswari College of Engineering, Bengaluru (Karnataka), India.
2Nischitha V N, UG Scholar, Department of Computer Science and Engineering, Raja Rajeswari College of Engineering, Bengaluru (Karnataka), India.
3Prakruthi SP, G Scholar, Department of Computer Science and Engineering, Raja Rajeswari College of Engineering, Bengaluru (Karnataka), India.
4Sindhu D K, UG Scholar, Department of Computer Science and Engineering, Raja Rajeswari College of Engineering, Bengaluru (Karnataka), India.
Manuscript received on 05 May 2019 | Revised Manuscript received on 17 May 2019 | Manuscript Published on 23 May 2019 | PP: 680-685 | Volume-7 Issue-6S5 April 2019 | Retrieval Number: F11190476S519/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Every year road accidents are getting increased. Somnolence and Drowsiness of drivers are one of the main reasons for road accidents. So we have to do more research on this area and find out novel technologies. Common exiting methods are vehicle based, behavioral based on physical parameters. And some techniques are effecting by disturbing the drivers. Methods need costly hardware and handling of data. Here, proposed method a driver sleepiness detection technique having low cost and high accuracy is developed. Here we are using image processing technique and Open CV, the drivers face is recording using webcam and face is detecting in every frames. Here aspect ratio of eyes(EAR), ratio of opening mouth (MOR) and length of nose ratio(NLR) are pointing. Then we calculate the facial landmarks using dlib library. Detection of drowsiness is depends on the computation of EAR, MOR and NLR and threshold values. Machine learning algorithm haar-cascades algorithm is used.
Keywords: Somnolence, Webcam, EAR, MOR, NLR, Pedestrian Walking.
Scope of the Article: Computer Vision