Font Size: a A A

Research On The Key Technology Of Close Range Photogrammetry For Location Of Prefabricated Buildings

Posted on:2023-02-17Degree:MasterType:Thesis
Country:ChinaCandidate:T F SuiFull Text:PDF
GTID:2530307037489854Subject:Photogrammetry (Professional Degree)
Abstract/Summary:PDF Full Text Request
In recent years,China is vigorously promoting the prefabricated building,but China’s prefabricated building started late,and the industry standard of prefabricated building is not unified,resulting in a series of problems such as too many prefabricated component factories,unreasonable production,poor quality and so on.Especially in the construction process,due to the lack of professional knowledge of workers,there are problems such as inaccurate positioning,long positioning time and mismatch between accessories in the site construction of prefabricated buildings,which greatly reduces the quality of prefabricated buildings and brings safety risks.In order to improve the quality and efficiency of prefabricated building construction site positioning,this paper innovatively introduces close-range photogrammetry technology into prefabricated building construction positioning,and carries out real-time and high-precision positioning of prefabricated building structures(fittings)by photographing prefabricated building structures(fittings).The use of close-range photogrammetry to prefabricated building structures(fittings)positioning,while ensuring accuracy,improve the efficiency of prefabricated building construction,promote the development of prefabricated building to modernization.Close shot photogrammetry prefabricated building structure(fittings)a positioning mainly includes three aspects: first,the prefabricated building structure(fittings)pieces of accurate identification,prefabricated building structure(fittings)a accurate identification and classification is the first step in the close shot photogrammetry and most important step in this step,the main use of deep learning target recognition,the method can identify the target in the complex environment;Second,accurate extraction of assembly building structures(fittings),accurate extraction of assembly building structures(fittings),the quality of contour extraction directly affects the quality of photogrammetry positioning;Third,the accurate positioning of the assembly building structure(fittings),the key points of the assembly building structure(fittings)into the relative coordinate system of the assembly building coordinate system,and in the way of images,intuitive performance of the component’s moving direction and moving distance.Paper in close shot photogrammetry real-time positioning prefabricated building structure(fittings)a location,real-time image through the camera,the use of deep learning approach to image the prefabricated building structure(fittings)pieces of identification,access to prefabricated construction structure box(fittings)a prediction to predict box normalization processing can avoid error identification and will predict box enlarge.Edge detection is carried out in the prediction frame,and the combination of deep learning and traditional edge detection is used to ensure the efficiency and quality of edge extraction.By using shape matching,the accurate contour of assembled building structures(fittings)is obtained.Through coordinate transformation,the coordinates of assembled building structures(fittings)are transformed into coordinates in the relative coordinate system of the construction site to complete the real-time positioning of photogrammetry.Public housing project in He Fei TIANMENHU lake as an example,and combined with laboratory experiments,through the way of photogrammetry,prefabricated building structure(fittings),real time calculation of prefabricated building structure(fittings)pieces of the key points in the coordinate,gets the coordinates of prefabricated building components with ruler to measure the coordinates of the comparison,the error is less than 3 mm,satisfy the requirement of the prefabricated construction precision.In addition,video is shot with photogrammetric cameras at a frame rate of about 30 FPS,which meets basic visualization requirements.Through this experiment,it is concluded that the close range photogrammetry is superior to the traditional total station measurement in positioning quality and efficiency,which proves that the close range photogrammetry technology can be used as the construction positioning of prefabricated buildings.
Keywords/Search Tags:prefabricated building, close-range photogrammetry, object identification, edge detection, shape matching
PDF Full Text Request
Related items