Eye Fatigue Algorithm for Driver Drowsiness Detection System
This paper proposed an algorithm that used the combination of the Viola-Jones technique, Circular Hough Transform (CHT), Histogram Equalization, Canny Edge detection and percentage of eyelid closure (PERCLOS) technique to accurately detect the eyes condit
- PDF / 2,350,400 Bytes
- 15 Pages / 439.37 x 666.142 pts Page_size
- 16 Downloads / 195 Views
Faculty of Engineering, Technology & Built Environment, UCSI University, Kuala Lumpur, Malaysia [email protected], {leonghy,MohdRizon}@ucsiuniversity.edu.my 2 Heriot Watt University Malaysia, Putrajaya, Malaysia [email protected]
Abstract. This paper proposed an algorithm that used the combination of the Viola-Jones technique, Circular Hough Transform (CHT), Histogram Equalization, Canny Edge detection and percentage of eyelid closure (PERCLOS) technique to accurately detect the eyes condition of the driver. This proposed algorithm achieved a 92.5% accuracy of eyes open images detection and 86.65% accuracy of eyes close images detection with 50 samples each. For the real-time video processing, it achieved 90% accuracy during daytime and 86% accuracy during night-time for the eyes drowsiness detection. Keywords: Drowsiness detection Viola-Jones Circular Hough Transform Image processing Eyes detection Eyes state analysis
1 Introduction In every country, curbing road accidents have always been one of the utmost priorities as they contributed highly to causing high damage, even involving death accidents. Based on the data from the Traffic Investigation and Enforcement Department, Bukit Aman [1], total road accidents in Malaysia has increased from 373,071 to 533,875 in the last decade (2008–2017). In a research study, human behaviour contributes about 95% of the factors that causes road accident [3]. According to The Star newspaper, three main causes of road accidents are the poor attitude of drivers, drivers using mobile phones while driving and driver’s drowsiness [2]. The poor attitude of drivers and using mobile phones while driving are considered as personal behaviour that hardly controlled or managed by a third party. The only way to prevent an accident comes from the driver’s self-initiative. To overcome driver drowsiness, a few researchers proposed an intelligent algorithm to alert the driver and prevent accidents to happen. The drowsy driver must be alert and asked to take a short nap after some key symptoms such as ‘difficulty to keep the eyes open’ and ‘heavy eyelids’ that are experienced by the driver [5]. The driver awareness is monitored using the determined key symptoms of drowsiness. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 J. I.-Z. Chen et al. (Eds.): ICIPCN 2020, AISC 1200, pp. 638–652, 2021. https://doi.org/10.1007/978-3-030-51859-2_58
Eye Fatigue Algorithm for Driver Drowsiness Detection System
639
2 Related Work Some efforts have been reported in the literature on the development of driver drowsiness detection system. Various approaches and techniques were proposed. Prakash Choudhary [6] performed a series of surveys for various drowsiness detection techniques including the eyes blinking based technique, yawning based technique, PERCLOS based, template matching technique, artificial neural network based, EEG based technique and vehicular based. However, the EEG based technique and vehicular based methods were then
Data Loading...
