Evaluation of Hand Washing Procedure Using Vision-Based Frame Level and Spatio-Temporal Level Data Models

Hand hygiene is obligatory for all healthcare workers and vital for patient care. During COVID-19, adequate hand washing was among recommended measures for preventing virus transmission. A general hand-washing procedure consisting several steps is recommended by World Health Organization for ensuring hand hygiene. This process can vary from person to person and human supervision for inspection would be impractical. In this study, we propose computer vision-based new methods using 12 different neural network models and 4 different data models (RGB, Point Cloud, Point Gesture Map, Projection) for the classification of 8 universally accepted hand-washing steps. These methods can also perform well under situations where the order of steps is not observed or the duration of steps are varied. Using a custom dataset, we achieved 100% accuracy with one of the models, and 94.23% average accuracy for all models. We also developed a real-time robust data acquisition technique where RGB and depth streams from Kinect 2.0 camera were utilized. Results showed that with the proposed methods and data models, efficient hand hygiene control is possible.

Face Mask Identification Using Spatial and Frequency Features in Depth Image from Time-of-Flight Camera.

Face masks can effectively prevent the spread of viruses. It is necessary to determine the wearing condition of masks in various locations, such as traffic stations, hospitals, and other places with a risk of infection. Therefore, achieving fast and accurate identification in different application scenarios is an urgent problem to be solved. Contactless mask recognition can avoid the waste of human resources and the risk of exposure. We propose a novel method for face mask recognition, which is demonstrated using the spatial and frequency features from the 3D information. A ToF camera with a simple system and robust data are used to capture the depth images. The facial contour of the depth image is extracted accurately by the designed method, which can reduce the dimension of the depth data to improve the recognition speed. Additionally, the classification process is further divided into two parts. The wearing condition of the mask is first identified by features extracted from the facial contour. The types of masks are then classified by new features extracted from the spatial and frequency curves. With appropriate thresholds and a voting method, the total recall accuracy of the proposed algorithm can achieve 96.21%. Especially, the recall accuracy for images without mask can reach 99.21%.

A Depth Camera-based Warning System Design for Social distancing Detection

Nowadays, COVID-19 is raging around the world. Because of its highly contagious, people have to take many measures and change their daily lifestyle to face it. Keeping social distancing is particularly important for the prevention of COVID-19, especially for the administrator of public spaces, it makes sense to urge people to maintain social distancing. However, if the administrator directly carries out social distancing management, it will consume a lot of manpower and material resources, it is necessary to design a system that can automatically detect social distancing status in the areas. For the studies of social distancing detection, most of the related works use pixel analysis techniques based on images to obtain distance data, but this type of technique may produce large errors due to the difference camera angles. Therefore, in this paper, we plan to present a design of the social distancing detection and warning system by using the devices of high-precision depth camera and Android-based smart glasses. By using the depth camera, we can obtain the distance data more accurately to prevent misjudgment due to insufficient information acquisition, in addition, the using of smart glasses as the information terminal in order to provide relevant social distancing warning information to the area administrators more quickly and accurately. This system will not only benefit area administrators directly, but will also provide the basis for research in the area of social distancing risk in public places. © 2021 IEEE.

Design of Contact Force for Ultrasonic Scanner

The global hospital's capacity can gradually decline, especially after the outbreak of the COVID-19, and thus traditional medical methods can no longer bear a large number of patients. At present, most hospitals rely on doctors and nursing staff to diagnose and treat patients. This not only increases the burden on doctors and nursing staff but also greatly reduces the burden on health care quality. In order to obtain better health care quality, automation is one of the important factors in solving medical quality problems. We are conducting automated introduction research for the ultrasonic scanner. Robotic arms are used to replace doctors for consultations by adding jelly and injection buttons to the robotic arm. In terms of the contact between the end of the robotic arm and the human body, we introduced the force sensor and the depth camera into the robotic arm. With the force sensor and the depth camera feedback data, we perceive the feedback of the ultrasonic scanner and the human body contact force. The results show that our design can greatly increase the amount of hospital's capacity and reduce the burden on doctors. © 2021 IEEE.