Font Size: a A A

Research On Human Posture Recognition System Based On Accelerometer

Posted on:2015-02-18Degree:MasterType:Thesis
Country:ChinaCandidate:Y X XueFull Text:PDF
GTID:2268330428469196Subject:Measuring and Testing Technology and Instruments
Abstract/Summary:PDF Full Text Request
It’s a new technology that human posture recognition based on the accelerationsensor and the monitoring technology. The human posture recognition system has a lot ofadvantages, such as the easy detection, non-affected by the environment and spacerestrictions, people’s privacy protection, and so on. So it is easy to be accepted by thepeople. Its hardware platform adopts Android mobile with accelerometer. Its algorithmuses time-domain and frequency-domain algorithm. It can phone automatically and sendmessages to the guardian to get relief in emergency situations.1) Optimization of hardware platformThe previous hardware platform was based on single-chip computer. The sensors inthe device were placed on the different parts of the body to obtain appropriate data andidentify the human posture. To optimize the device, Android mobile with accelerometer isadopted as the hardware platform. It solves the questions of the power supply and fixedplace of the wearable sensor device.2) Human posture data acquisition and improvement measuresThe data collection procedure saves data to the mobile memory. It confirms that thehuman posture recognition and monitoring system based on the Android platform isfeasible.3) Research method of time domain and frequency domainBased on the acceleration modulus value, the human posture recognition methodimproves the recognition rate. But to use the time domain method only may lead to theerror and mistake of the posture recognition. In order to improve the human posturerecognition accuracy, the system analyses the signals for the different postures in thefrequency domain.4) Software design and implementationThe feature of the graphical user interface in Android mobile and the functions ofphone, message, GPS and network location are realized by using of the Eclipse integrateddevelopment tool. The system software to identify and monitor the human posture iscompleted.
Keywords/Search Tags:Posture Recognition, Accelerometer, Android, Acceleration Modulus Value, Monitoring Technology
PDF Full Text Request
Related items