Driver Drowsiness Detection Based on Face Feature and Perclos

Driving vehicles are complex and require undivided attention to prevent road accidents. Fatigue and distraction are a major risk factor that causes traffic accidents, severe injuries, and a high risk of death. Some progress has been made for driver drowsiness detection using a contact-based method that utilizes vehicle parts (such as steering angle and pressure on the pedal) and physiological signals (electrocardiogram and electromyogram). However, a contactless system is more potential for real-world conditions. In this study, we propose a computer vision-based method to detect driver's drowsiness from a video taken by a camera. The method attempts to recognize the face and then detecting the eye in every frame. From the detected eye, iris regions for left and right eyes are used to calculate the PERCLOS measure (the percentage of total time that eye is closed). The proposed method was evaluated based on public YawDD video dataset. The results found that PERCLOS value when the driver is alert is lower than when the driver is drowsy.

Contactless System with Mask and Temperature Detection

Coronavirus disease in 2019 has affected the world very badly on a large scale. One of the important protection methods is to wear masks in public areas. Also, while using public services it is important to wear a mask correctly if you want to use their services. However, there is very few researches on face mask detection based on image analysis. In this paper, we propose Face Mask, which is a high-accuracy and efficient face mask detector. The proposed system is a one-stage detector, which consists of a pyramid network to fuse high-level semantic information with multiple feature maps, and a module to focus on detecting face masks. In addition, we also propose a novel cross-class object removal algorithm that will reject predictions with low confidences and the high intersection of the union. Besides, we also focus on the possibilities of implementing Face Mask with a light-weighted neural network MobileNet for embedded or mobile devices. In this paper, we introduce an affordable solution aiming to increase COVID-19 indoor safety, covering relevant aspects: 1) contactless temperature sensing 2) mask detection. Contactless temperature sensing subsystem relies on Arduino Uno using an infrared sensor or thermal camera, while mask detection is performed by leveraging computer vision techniques and Deep Learning Techniques.

Wrist Vascular Biometric Recognition Using a Portable Contactless System

Human wrist vein biometric recognition is one of the least used vascular biometric modalities. Nevertheless, it has similar usability and is as safe as the two most common vascular variants in the commercial and research worlds: hand palm vein and finger vein modalities. Besides, the wrist vein variant, with wider veins, provides a clearer and better visualization and definition of the unique vein patterns. In this paper, a novel vein wrist non-contact system has been designed, implemented, and tested. For this purpose, a new contactless database has been collected with the software algorithm TGS-CVBR®. The database, called UC3M-CV1, consists of 1200 near-infrared contactless images of 100 different users, collected in two separate sessions, from the wrists of 50 subjects (25 females and 25 males). Environmental light conditions for the different subjects and sessions have been not controlled: different daytimes and different places (outdoor/indoor). The software algorithm created for the recognition task is PIS-CVBR®. The results obtained by combining these three elements, TGS-CVBR®, PIS-CVBR®, and UC3M-CV1 dataset, are compared using two other different wrist contact databases, PUT and UC3M (best value of Equal Error Rate (EER) = 0.08%), taken into account and measured the computing time, demonstrating the viability of obtaining a contactless real-time-processing wrist system.

Methodology for GD&T verification in an innovative benchmark part for contactless scanning systems

This paper presents an innovative method to investigate the accuracy and capability of contactless laser scanning systems in terms of geometrical dimensioning and tolerancing (GD&T) control. The current work proposes a standard benchmark part with typical features conforming to different families of GD&T. The benchmark part designed consists of various canonical features widely used in an engineering and industrial applications. Further, the adopted approach includes the methodology for comparison of geometry using a common alignment method for contactless scanning system and a CMM. In addition, proposal of different scanning orientation methods for contactless system is also realized. Surface reconstruction of the benchmark model is achieved using different reverse engineering software, and results are analyzed to study the correlation between different geometries of contact and contactless system. Considering the contact based measurement as a reference, different models developed were analyzed and compared in terms of geometrical and dimensional tolerance. The proposal of standard benchmark part and methodology for GD&T verification will provide a simple and effective way of performance evaluation for various contactless laser-scanning systems in terms of deviations.

Measurement of the steady state tearing in thin sheets using the contactless system

The purpose of the present paper is the diagnostic method of the steady state tearing tests in thin steel sheets. The camera-based contactless system was used in the experimental research. The MATLAB software was used for the processing of real-time images. The data collected in the experiments were used to establish the R-curves in terms of the CTOD (crack tip opening displacement) based resistance curves. The innovative techniques allow extracting a large amount of geometrical information about the entire tearing process and the crack tip development. The steady state tearing was evaluated using automotive steel sheets and the tearing resistance was determined for three grades of thin sheets. The paper describes a comprehensive method of obtaining the material properties necessary for modelling and simulation. The proposed methodology for the testing and evaluation of the steady state tearing is applicable to thin sheets. The measurement method within the test for the determination of the tearing resistance was based on the image analysis.

Opioid overdose detection using smartphones

Early detection and rapid intervention can prevent death from opioid overdose. At high doses, opioids (particularly fentanyl) can cause rapid cessation of breathing (apnea), hypoxemic/hypercarbic respiratory failure, and death, the physiologic sequence by which people commonly succumb from unintentional opioid overdose. We present algorithms that run on smartphones and unobtrusively detect opioid overdose events and their precursors. Our proof-of- concept contactless system converts the phone into a short-range active sonar using frequency shifts to identify respiratory depression, apnea, and gross motor movements associated with acute opioid toxicity. We develop algorithms and perform testing in two environments: (i) an approved supervised injection facility (SIF), where people self-inject illicit opioids, and (ii) the operating room (OR), where we simulate rapid, opioid-induced overdose events using routine induction of general anesthesia. In the SIF (n = 209), our system identified postinjection, opioid-induced central apnea with 96% sensitivity and 98% specificity and identified respiratory depression with 87% sensitivity and 89% specificity. These two key events commonly precede fatal opioid overdose. In the OR, our algorithm identified 19 of 20 simulated overdose events. Given the reliable reversibility of acute opioid toxicity, smartphone-enabled overdose detection coupled with the ability to alert naloxone-equipped friends and family or emergency medical services (EMS) could hold potential as a low-barrier, harm reduction intervention.