Smart Machinery

A Vision Based Biosignal Health Management System and Its Applications to Driving Safety

Principal Investigator:Bing-Fei Wu, Ph.D. of National Chiao Tung University
Abstract

“Health Management System Using Image based Physiological Signals for the Application of driving safety” is a complete solution for the safety of transportation. With the technology of computer vision, the system can effectively monitor the status and behavior of drivers by face detection and recognition, image based physiological information measurement, behavior recognition, fatigue detection, drunk driving recognition, and painful expression analytics. If the detected status is abnormal, the system is going to take some suitable actions, warning the driver in real-time, or sending the messages to authority concerned.

    • Contactless measurement: the measurement is to detect the flow change of facial blood by imaging, then retrieves the heart rate value. This method not only reduces the user's inconvenience of remembering to wear a bracelet and inaccurate interference during exercise, and the measurement is more flexible.
    • Physiological measurement: By a single camera, our device can identify the driver, and keep tracking the face and detecting the behavior to analyze the driving and physiological status.
    • Multi-angle measurement: Unlike ordinary face recognition system which needs the face fixed in front of the camera for a few seconds, our technology of face recognition system has no limit for the face angle, and can be more competitive and applied on more aspects.
1learn more
Team Introduction 

We are e-Bus, and we are committed to make everyone go home safe and sound. The "eBus" product is developed since we would like to provide a solid smart armor to our customers. This system can effectively avoid dangerous driving behaviors, reduce the number of car accidents, and enhance the safety quality of public transport.

In the future, eBus team will enhance its own competitive advantage, and develop towards different applicable aspects. We also expect to expand overseas markets, develop customer base, look for cooperation opportunities in other fields by the 2018 CES exhibition which led by Ministry of Science and Technology. Our effort and various of assistance can be the cornerstone to our team and make us become Taiwan's unicorn company.

Entry Barrier

With respect to the products of heartbeat calculation by continuous face images in the market, the commonly used core algorithm is only able to resist the noise caused by face shaking. There are still some limits for the application, for example, when applied to the heartbeat detection of drivers, there is serious image noise influenced by uneven shadows on face. The motion-tolerant algorithm still cannot overcome the difficulty. Our technology not only overcome the interference of facial motion, further we also reduce the influence of the noise generated by the uneven lighting of the face.

In addition, our technology combines with the triple system function of heartbeat detection, blood pressure and heart rate variation detection by facial image. The integrated three kinds of physiological information measurement by image detection is the most unique in the world now. Currently, no products utilize the contactless methods to detect and integrate various physiological information in the market.

The innovative features of our technology are as follows:

    1. The first camera base blood pressure measurement system in the world.
    2. A single camera can measure heartbeat, blood pressure, and heart rate variability.
    3. 90-degrees angle for face detection technology and real-time processing of the technology of arbitrary motion face recognition.
    4. Image heartbeat measurement system: the heartbeat value can be measured as high as 180 or more.
    5. Big data of cloud analysis of individual physiological conditions and the artificial intelligence analysis for health status.
Market Scope

In 2011, the United Nations announced the "Ten-Year Plan" that hopes it will call on all countries in the world to pay attention to road safety issues and devote to enhance the transport system from 2011 to 2020. Due to the announcement of United Nations and our government policy for prevention and control of traffic accidents, the traffic safety industry has high growth potential in Taiwan and foreign countries. We hope that we can take this trend and make a contribution to traffic safety. Therefore, our team researched, analyzed the causes of car accidents and then propose a set of image-based driver health and safety monitoring system. We expect to detect the driver's physiological and behavioral conditions immediately before the traffic accident happen and alert the drivers for providing a guarantee to the driver's life and property safety.

 In the initial stage, the operation plans focuses on the stability, application, and technical transfer of the core technology. At the same time, it carries out long-term testing of products in the field of vehicles and collects a large amount of data to establish an analysis database.

 In the medium term, our team will exert its own strengths to make breakthroughs in technology research and development. In the latter stage, we expect to transition to an image-based AI technology company. Through the pre-hard work and effort, our company will be positioned as a comprehensive analysis of large artificial intelligence technology providers. Also we will develop deeper to make the core technology more flexible and permeate into the different scope applications in healthy business, fitness and entertainment markets. Through these innovations and services, it is expected to gain great potential of profit.

Video Introduction