Font Size: a A A

Development Of Comprehensive Management Integration Platform Of Physical And Technological Prevention And Research On Face Recognition

Posted on:2018-10-01Degree:MasterType:Thesis
Country:ChinaCandidate:E P WangFull Text:PDF
GTID:2348330542470394Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
With the economic,social and environmental improving constantly,a pumped storage power station has been rapid development.As the pumped storage power station area is generally at least 50 kilometers continuous,forest fire,reservoir to prevent staff drowning,dam to prevent people engaged in damage to the safety management than the general thermal power plant is more important.For the safety of almost all of the pumped storage power stations successively set up subsystems of physical and technological prevention,these subsystems are usually different manufacturers at different times using different databases and software architecture development,resulting in independent subsystems,information sharing difficulties.In order to protect the investment,the existing information subsystem can not be discarded.Thus,the paper studies the physical and technological prevention system of a pumped storage power station and its related problems,to design and implement of an physical and technological prevention system integration platform.The platform integrates the existing physical and technological prevention subsystems of enterprise adds some self-developed function software,provides unified command and dispatching ability,implements unified portal?unified rights management?unified information display and all information subsystems cooperating jobs.The linkage between the physical and technological prevention subsystems can effectively improve the accuracy of the alarm and the reliability of the control action.In addition,aiming at the problem of wide geographical and human monitoring in enterprise video surveillance system,a face monitoring technology is proposed to assist the video monitoring scheme.The face detection algorithm based on AdaBoost and the algorithm based on ASM feature extraction are improved and a face recognition of auxiliary video surveillance software module is developed,which can effectively assist the identification of characters,enhance the level of enterprise safety management.Firstly,a detailed investigation is carried out on the status of the physical and technological prevention systems of the pumped storage power station.The functional requirements and difficulties of the system are analyzed.On this basis,the overall architecture and function are designedSecondly,the face detection problem is studied.In the face detection algorithm based on AdaBoost,in the analysis of sample selection?AdaBoost algorithm training process?design of cascade classifier and detection process,we make the corresponding improvement to weighted parameters of weak classifier problem and the correlation problem of the traditional AdaBoost algorithm.The experiental results show that the new method will increase the detection rate and improve the detection speed.Then,the problem of face feature extraction is discussed in face feature comparison.Feature location is an important step in face feature extraction algorithm based on active shape model.By the pupil position which is positioned by matching the eye positioning algorithm based on the R channel,the traditional ASM initialization algorithm is improved.The experiental results show that compared with the traditional ASM algorithm,the improved ASM algorithm is more accurate.Finally,this paper describes the implementation details of the system,using SQL Server as the database management system?Power Designer as the system database modeling tool?C#as the foreground program design language?Matlab as intelligent algorithm development language,Through the formal programming thought and the guidance of the principle,completing the development of comprehensive management Integration Platform of physical and technological prevention.
Keywords/Search Tags:integrated system, face detection, face feature extraction, Active shape model
PDF Full Text Request
Related items