Human Motion Recognition And Comprehensive Analysis System

In this paper, a new method was presented to realize motion detection on a mobile device.The scheme can recognize the people’s motions state according to the acceleration data as long as they simply carry a mobile device with a build-in triaxial accelerometer. The features of the motion signal are extracted in frequency domain and time domain using the method of comprehensive analysis. To enhance the adaptability of the method, the algorithm of independent direction of mobile device algorithm has been applied. The 11 major components, which have greatest contribution to the motion detection, are selected from the 21 motion’s features by principal component analysis, so reduced the input dimensions and decreased computational complexity of time and space of the algorithm. Based on the analysis and synthetic comparison of various classification algorithm, the J48 decision tree is accepted. According to the characteristics of the people nature motion, the hidden Markov model was introduced to improve the detection accuracy. Experiments, with different person and different motion, show that the synthesis algorithm has good accuracy and adaptability, the highest recognition rate achieved 96.13%.This paper was presented a method for calculating the exercise intensity based on user’s age,sex, weight, height and other factors. This method can be better highlight the strength of user’s exercise intensity under different physical characteristics. After collecting the user’s some latest exercise intensity data in few days. Then this paper will be create a exercise intensity model for the user totally based on user’s individual characteristics. And then according to this model, we will recognize the user’s all abnormal states of motion. At last we will warn these motion states to the user. It is proved by experiments that this abnormal motion detection model can accurately determine the user motions, so that users will have a more healthy and comfortable life in scientific way.Finally, two different applications are developed, based on the theory of this paper. The one is a full-featured mobile phone application, in order to achieve the user motions recognition. This application can done a good job in the motion recognition, collect and calculate the amount of exercise, show the user specific movement trajectory and interact with servers, and some other functions. In order to make the motion data to became a share resources for users, and improvecommunication with others, the application adds social sharing function.The other application is remote sever system. Remote server can collect and analysis large numbers of users’ motion data. Then depending on the user different gender, age, height and weight, according to the user recently motion data. The server can model the motion intensity for every user. This model can identify the abnormal motions of the user. If user under the abnormal condition, the program will warn the user to prevent them from excessive exercise or too little exercise.