Saturday, September 12, 2015

Accident Detection System IJSRD

Journal Details : IJSRD - International Journal for Scientific Research & Development| Vol. 2, Issue 02, 2014 | ISSN (online): 2321-0613

Title : Accident Detection System

Author Name : Aswathy James,Nimmy Rose Sam ,Neena K Pius, Jito Tomy Joseph

College Name : Amal Jyothi College of Engg. (Mahatma Gandhi University) Kerala, India

Accidents like falls in people can happen due to reasons like old age, unconscious mind and poor motor control. Such accidents may result in broken or fractured bones, cuts, and even death. Living with the fear of falling leads to isolation, worsening of mental health, and general degradation of quality of living. If the accidents cannot be prevented, the next best option is detecting them accurately, and alerting the required authorities for medical help. This paper introduces an accident detection mechanism with the help of the Smartphone. Smartphone’s gather the information using the Built-in tri-accelerometer and the data obtained from the accelerometer are analyzed using the displacement based algorithm. On detecting an accident, the user is prompted to respond and can use touch or voice recognition to reduce the false positives. If the user is not responding within a period of time, the application generates alerts like emails and text messages and alerts the predefined contacts. If any unconscious situations occur, the user can press the panic button and the appropriate message/audio/video/image will be sent to the predefined contacts.The system also provides the hospitality facility.That is by providing the details of the nearest hospital.

EXISTING SYSTEM AND DRAWBACKS:

Current Methods for Automatic Fall Detection includes the following. Falls that occur in isolation, immediate help by detecting it quickly and accurately is an important factor. 

For this purpose Personal Emergency Response Systems which will detect the fall by generating alerts. It will also reduce the time between the fall and the arrival of the medical attention. PERS uses pendant for the alert generation but it will not work if the person is unconscious. 

In order to classify the fall automated fall detecting systems use one of the three methods which include Vibration recognition,image recognition,worn devices. 

Vibration recognition: It contains a device which is inplanted on the groundor floor in order to monitor sound or the vibration.By comparing the vibration frequency it is easier to identify the fall. 

Image recognition: This is done by placing a camera in a fixed location and the movement patterns are recorded.The system adapts to the locations in which a single human enters/exits the room and remains inactive (lying/sitting on bed/chair). Common paths from entry points to inactive areas are then traced and remembered. It suspects a fall if a person becomes inactive in the middle of a common path. It have the disadvantages such as if the person is outside the monitored area the system become ineffective. 

Worn Devices: In this the user need to wear external sensors and it will uses a tri-axial accelerometer. If a specific pattern or threshold is broken, the device alerts a wireless receiver, which then alerts emergency contacts. It also faces some problems such as inconvenience and affects the daily activities of the life. 

Recognition by acceleration threshold: This system detects fall condition by analyzing acceleration data from an accelerometer. If the amplitude of the acceleration crosses the lower and upper thresholds and if there is a change in position then fall condition is detected. 





No comments:

Post a Comment