Font Size: a A A

Design And Implementation Of The Intelligent Calland Search System For Firefighters

Posted on:2020-12-23Degree:MasterType:Thesis
Country:ChinaCandidate:D H MengFull Text:PDF
GTID:2392330578967070Subject:Engineering
Abstract/Summary:PDF Full Text Request
With the development of society and the progress of science and technology,people's living standard has been greatly improved.Due to human negligence or natural factors,fire always accompanies human life and causes great material damage and casualties to human beings.Therefore,in order to reduce material damage and provide timely relief to people in danger of fire so as to reduce casualties,the society has a growing demand for firefighters.In order to cope with the needs of the society,the needs of firefighters are increasing.Because the fire is merciless,it often brings risks for life of the firefighters who are in the front line of battle,and causes the unanticipated loss.With the continuous occurrence of fire,the life safety of firefighters is particularly important.Therefore,in recent years,researchers have begun to conduct in-depth research in the field of intelligent fire-fighting life-saving equipment,and it is possible to apply the posture recognition technology of firefighters to the intelligent fire-fighting equipment.In this thesis,the human body posture recognition technology is introduced in detail.It begins from the basic movement of human body(static,walking and running,etc.)to analyze the internal relationship between different human body posture and acceleration,and extracts the characteristic values of human body combined acceleration under different posture.And then after the introduction of the machine algorithm of gesture recognition,the support vector machine(SVM)algorithm and random forest after optimization of support vector machine algorithm(RF-SVM)are used respectively for the simulation of the human body posture in the MATLAB environment.The comparison results show that the optimized RF-SVM algorithm has a certain robustness,avoids the over-fitting phenomenon,and achieves high recognition accuracy,so that the expected effect has been obtained.Firefighters intelligent help-calling and search system is based on the STM32 module for the overall architecture,with the logic combination of power supply module,display module,positioning module,MPU6050 acceleration module,and wireless transmission module.Finally the results of hardware implementation are sent to the upper machine(PC)by means of wireless transmission,so as to judge whether the firefighters are in dangerous situation.
Keywords/Search Tags:Gesture recognition, Combined acceleration eigenvalues, PC, The SVM, The RF-SVM
PDF Full Text Request
Related items