Real-Time Vision Based Driver Drowsiness Detection Using Partial Least Squares Analysis

Selvakumar, K.; Jerome, Jovitha; Rajamani, Kumar; Shankar, Nishanth

doi:10.1007/s11265-015-1075-4

Cited by 13 publications

(12 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A complete example of a drowsiness detector based on eye closure is found in [83]. This study offers a full pipeline description from image capture to percentage of eye closure (PERCLOS) computation.…”

Section: ) Algorithms Structurementioning

confidence: 99%

Resource-Constrained Machine Learning for ADAS: A Systematic Review

Borrego-Carazo

Castells-Rufas

Biempica³

et al. 2020

IEEE Access

View full text Add to dashboard Cite

The advent of machine learning (ML) methods for the industry has opened new possibilities in the automotive domain, especially for Advanced Driver Assistance Systems (ADAS). These methods mainly focus on specific problems ranging from traffic sign and light recognition to pedestrian detection. In most cases, the computational resources and power budget found in ADAS systems are constrained while most machine learning methods are computationally intensive. The usual solution consists in adapting the ML models to comply with the memory and real-time (RT) requirements for inference. Some models are easily adapted to resource-constrained hardware, such as Support Vector Machines, while others, like Neural Networks, need more complex processes to fit into the desired hardware. The ADAS hardware (HW platforms) are diverse, from complex MPSoC CPUs down to classical MCUs, DPSs and application-specific FPGAs and ASICs or specific GPU platforms (such as the NVIDIA families Tegra or Jetson). Therefore, there is a tradeoff between the complexity of the ML model implemented and the selected platform that impacts the performance metrics: function results, energy consumption and speed (latency and throughput). In this paper, a survey in the form of systematic review is conducted to analyze the scope of the published research works that embed ML models into resource-constrained implementations for ADAS applications and what are the achievements regarding the ML performance, energy and speed trade-off. INDEX TERMS Machine learning, embedded software, automotive engineering, GPU, FPGA, ADAS. I. INTRODUCTION Safety is increasingly important for drivers and other users of the driving environment such as pedestrians, cyclists, bikers, scooters, etc. Only in 2016, road accidents in the European Union summed up to 25.600 fatalities and 1.4 million people injured [1]. Many measures are being taken by public institutions (improving road quality, enhancing driver consciousness and driving regulations, etc.) to improve these numbers. However, the cause of most of these accidents resides in human errors or distractions. For this reason, automotive firms are pushing up research on improving safety. The cornerstone of these advances are Advanced Driver Assistance Systems (ADAS). ADAS systems consist in several sensors, processing and actuators components that help drivers to avoid accidents and The associate editor coordinating the review of this manuscript and approving it for publication was Ye Duan.

show abstract

Section: ) Algorithms Structurementioning

confidence: 99%

Resource-Constrained Machine Learning for ADAS: A Systematic Review

Borrego-Carazo

Castells-Rufas

Biempica³

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Many different approaches have recently been proposed to detect driver drowsiness. In current studies, various techniques such as artificial neural networks (Sayed and Eskandarian, 2001), SVM and the Naive Bayes Classifier, fuzzy logic (Senaratne et al, 2007;Bergasa et al, 2006), Gabor filter (Flores et al, 2010), PCA and the Kalman filter (Selvakumar et al, 2015) have been employed for drowsiness detection. Most research in this group is conducted by studying the face and the eyes and extracting the signs and features of drowsiness from them.…”

Section: Problem Definition and Research Objectivesmentioning

confidence: 99%

“…When driver's eyes are closed for a certain amount of time, which is determined as a threshold, the driver can be detected as being drowsy. Many scientists have conducted their research on PERCLOS algorithms which are very popular in many drowsiness detection approaches (Flores et al, 2010;Selvakumar et al, 2015;Jo et al, 2014). In these algorithms, the percentage of the eyes being open is measured in a certain period of time.…”

Section: Problem Definition and Research Objectivesmentioning

confidence: 99%

Robust real-time driver drowsiness detection based on image processing and feature extraction methods

Keyvanara

Salehi

Monadjemi

2018

IJVS

View full text Add to dashboard Cite

Recently, the human lifestyle has strongly been affected by the novel technological equipment. The applications of Artificial Intelligence are widely being utilised to improve the performance and quality of the modern life. One of the important applications of these techniques is to seek to improve public safety, including the safety of driving. The statistics indicate that the mortality of car accidents yearly constitutes a significant proportion of the overall deaths. A number of strategies have been studied to materialise driver drowsiness detection systems. One of the best strategies relies on image processing and computer vision methods. In this paper, a novel real-time method for driver drowsiness detection is presented. This method uses Haar wavelet-based features for face detection. The eye state determination has been performed using PCA feature extraction along with an SVM classifier. The proposed method has been implemented and tested on a real-time ARM based embedded system, with a camera installed in front of the driver. Results show that the presented intelligent system has a high detection accuracy, compared to the methods presented thus far, on the standard datasets such as BioID and RS-DMV.

show abstract

“…That study was able to detect the blink of an eye with an accuracy of 98.4%. While Selvakumar et al classified the eye conditions with Partial Least Squares (PLS) method after doing feature extraction with Local Binary Pattern Histogram (LBPH) [5]. The accuracy rate of eye classification expressed with a value of PERCLOS.…”

Section: Introductionmentioning

confidence: 99%

Modification on Brightness Enhancement for Simple Thresholding in Eyelid Area Measurement

Indrabayu

Tacok

Areni

2017

Proceedings of the 6th International Conference on Bioinformatics and Biomedical Science

View full text Add to dashboard Cite

This study aims to calculate eyelid area to classify eye condition become awake or drowsy as the initial stage of a drowsiness detection system. This work used 40x95 matrix of 500 single eye images from 4 respondents. The input image will be processed into brightness enhancement, grayscaling, thresholding and eyelid area measurement. The problem of thresholding process was to determine the optimal threshold value that can be used on images with different intensities. Brightness enhancement modification is proposed to equalize the intensity histogram value of the input image based on the reference image. Threshold value based on this modification can be used as optimal threshold value. With this optimal threshold value, the eye condition classification results show that average accuracy achieved 93.5%.

show abstract

Real-Time Vision Based Driver Drowsiness Detection Using Partial Least Squares Analysis

Cited by 13 publications

References 19 publications

Resource-Constrained Machine Learning for ADAS: A Systematic Review

Resource-Constrained Machine Learning for ADAS: A Systematic Review

Robust real-time driver drowsiness detection based on image processing and feature extraction methods

Modification on Brightness Enhancement for Simple Thresholding in Eyelid Area Measurement

Contact Info

Product

Resources

About