Real-time eyelid open/closed state recognition based on HLAC towards driver drowsiness detection

Abstract

This paper proposes a real-time eyelid state recognition method based on a video sequence. The human eye strongly reflects the mental state of an individual, such as attention, drowsiness, stress and confusion. In recent times, the automatic identification of such mental states using non-contact eyelid state recognition technology is proving to be a promising avenue for the development of such systems. In the field of Intelligent Transport Systems (ITS), high accuracy and real-time processing are necessary for detecting driver drowsiness in order to prevent accidents. To develop such a recognition method, we use Higher-order Local Auto-Correlation (HLAC). HLAC can represent the shape feature in images clearly without incurring a high computational cost, and it implements position invariability. A Support Vector Machine (SVM) is used to differentiate between open and closed eyelids on the basis of the HLAC feature. The proposed method can achieve open and closed eyelid recognition rates of 98.77% and 98.98%, respectively. We also verify the real-time processing capabilities of our method, thus confirming that it is effective for eyelid state recognition.

Driver Drowsiness Prediction Based on Multiple Aspects Using Image Processing Techniques

“…The majority of the research works are done based on the eye status of the driver [31][32][37] [41]. Drowsiness detection was done using LBPH [33] [48].…”

“…Wang and Qin [38] implemented a system based on the FPGA to detect the driver's drowsiness. Ishii et al [41] have proposed High-order Local Auto-Correlation (HLAC) for extracting the shape features and identified the attention, stress, drowsiness. Ling et al [42] has introduced a discriminative local feature vector for facial expression recognition using the sparse coefficients.…”

“…(x1,y1),(x2,y2)=(shape [37][0],shape [38][1]),(shape [40][ 0], shape [41][1]) Similarly, the computes of the mouth and facial cropped images were calculated in size. Detection of hand, when it is found on the face in an image that has been captured while yawning or nodding, is the prominent step in the present research work.…”

Dara

“…Across all the studies that deal with eye and head orientation, tracking is performed using a camera, and then, based on the collected images, a machine learning algorithm asserts the level of driver's attentiveness. In more detail, such systems were used to detect if drivers are looking at the road [1], or how tired they are [9,11]. In relation to hand tracking, the authors of [31,34,35] managed to assert drivers' activities using depth cameras.…”

Supervised Learning Techniques for Stress Detection in Car Drivers

“…There are various stress analysis approaches when we consider car driving scenarios as well. Many use physical characteristics, e.g., facial or eye images [18,19]. In others, physiological characteristics are considered [20][21][22].…”