IOS Platform Based Human Fall Monitoring System Design And Implementation

With the aging of progressively accelerating growth of the elderly proportion of the population, followed by increasingly prominent problems of the elderly,especially in the health problems of the elderly, and in many health problems threaten the elderly, the fall has become important factors. The elderly due to the fall did not get timely treatment of serious consequences caused by not only the elderly will cause severe physical and psychological trauma, but also to the social and individual families a heavy financial burden. Therefore, if we can design and implement a timely detect falls in older people and can alert the elderly to get timely relief fall detection software, not only can effectively reduce the risk of the elderly due to a fall caused by the disease, but also reduce the the elderly due to accidental falls and injuries to individuals and families economic burden on society.Based on the study and analysis of the current fall detection research at home and abroad on the basis of the design of a system to iOS platform based carrier triaxial acceleration sensor smartphone fall detection system. In this thesis, the acceleration SVM feature amount to human fall weightlessness, the body falls weightlessness time threshold, human fall amplitude SVM mean and human fall SVM peak multi-threshold classification joint judgment algorithm, first through a lot of experiments to obtain a variety of fall and daily activities of the acceleration data,and using MATLAB tool to draw a fall and daily activities of the graph to determine the appropriate human fall weightlessness, weightlessness human fall time, the magnitude of human fall and human fall SVM SVM mean peak threshold; the use of iPhone smart phone comes with accelerometer Get real-time human acceleration sensor data to calculate the acceleration SVM feature value, so that by the threshold determination algorithm successfully detected fall; after the fall of the system implementation uses the phone's GPS location detection capabilities fall into location,and location information falls sent via the mobile network to set the phone at the same time the guardian guardian dial phone, and uses cancel the alarm and active alarm mechanism reduces false positives and false negatives.Software system design is divided into three modules, parameter setting module,data processing module, falls intelligent alarm module. In the parameter settingmodule, the user can guardian contact information and message content to set up and editing, data processing modules using the iPhone built-in acceleration sensor data acquisition, and the median filtering the raw data smoothing, and finally in Analyzing a suspected fall under the SVM-based fall detection algorithm. Fall intelligent alarm module using the "one-button alarm", "manually cancel the alarm", "alarm" and warning mechanisms to improve the accuracy of fall detection to reduce false alarm rate.Fall detection system developed in the experimental verification stage experimental validation indicators, no-rules attitude fall group and daily activities of non-fall group experimental program to test two sets of experiments categories accuracy and false alarm rate fall detection system and the false negative rate.Finally, experimental verification results drawn from the analysis of the fall detection system sensitivity 89.3%, specificity 93.3% and accuracy was 92%.Through theoretical research and experiment, this system proved the accuracy of detection of falls among the elderly, and to achieve a fall after the alarm and positioning. The system detects a high accuracy, reliable performance and stability,but also the need for further improvement. System for the protection of the health of the elderly, reduce stress and promoting health is very broad application and research prospect of harmony has a positive meaning, and having the community.