Design And Implementation Of Pension Monitoring Software

The increasingly serious aging problem has become a frequent topic of conversation in the modern world,the deepening of the aging problem also promotes the rise of pension institutional.Institutional pension as one of the pension models,there are still exists the phenomenon of backward pension technology and inadequate pension service.With the rapid development of science and technology and the continuous research on the concept of elderly care,the use of modern technology and equipment to provide real-time and efficient intelligent elderly care services for the elderly in pension institutions has become an inevitable trend of social development.Based on this,this thesis designs and implements an elderly care monitoring software,which can detect the fall behavior of the elderly and monitor physiological parameters in real time.Through the intelligent analysis of behavioral and physiological parameters,the monitoring software provides services such as abnormal alarm,emergency rescue and emotional care,so as to comprehensively guarantee the physical and emotional health of the elderly.The main work includes:1.Investigate the research background and current situation of intelligent elderly care monitoring system at home and abroad,summarizes the existing problems of the current monitoring system.On this basis,conduct demand analysis from the perspective of monitoring software users,determine the final development scheme of the monitoring software,C/S and B/S hybrid architecture,Android operating system and Open CV visual algorithm library are used to develop the monitoring software.2.Comparison of common motion target detection algorithm,visual background extraction algorithm is chosen as the algorithm in this thesis,combined with external rectangular pixel histogram and shadow detection algorithm,to obtain the complete prospect target,and then select height to width ratio,body height,Hu moment feature and direction Angle of the four characteristics parameters of characterizing the prospect target behavior,finally,support vector machine was used to classify the fused characteristic parameters and identify the fall behavior.The simulation results show that the fall detect algorithm selected in this thesis meets the expected requirements.3.According to the actual needs,the monitoring software is divided into different functional modules,including system management,health monitoring,emergency rescue and other modules,and the design and implementation of each module are described.4.For the completed monitoring software,analyze common software testing schemes,the black box test and white box test are used to test software,and the test results are analyzed.The test results show that the monitoring software designed in this thesis meets the expected requirements.