Research On Measurement Technology Of Structural Construction Process Based On Machine Vision

Under the background of vigorously promoting infrastructure construction,in order to ensure the efficiency and safety of construction,construction departments at all levels have been paying attention to the development of construction monitoring technology.Among them,the construction measurement technology can provide effective data support for construction monitoring,and the quality of the measurement method will greatly affect the construction quality,project progress and even the safety of the structure and workers.In particular,the measurement methods in the construction process have attracted much attention.On the one hand,effective measurement methods can monitor the deformation or displacement of the structure during the forming process,and then the construction personnel can take corresponding control measures in a timely manner based on this,and prevent similar structures Collapse,casualties and other accidents;on the other hand,the automated measurement mode can not only quickly provide measurement data to speed up on-site construction progress and shorten the construction period of the project,but also reduce the demand for measurement personnel to ease the tension of engineering employment.status quo.With the great advances in optics and computer science,machine vision-based measurement methods can achieve automated and precise measurements,and their low equipment costs are well suited to the needs of construction sites.However,different from the previous application fields of machine vision measurement,the high complexity of the construction site and the variety of measurement requirements are the main reasons for limiting its widespread promotion.Relying on the installation project of a prefabricated viaduct,this paper proposes a set of automatic measurement methods based on the actual needs of the project,which helps to improve the speed and accuracy of on-site construction measurement.For the position detection of reserved steel bars in prefabricated structures,this paper proposes an automatic method for measuring the position of steel bars by combining the image-based large-field-of-view parallel plane measurement method and deep learning target detection.In order to evaluate the accuracy and feasibility of the method,this study determined the parameters affecting the variation of the error based on the theoretical analysis of the measurement method.The measurement accuracy and effect are verified by the test method of indoor control variables.Finally,the method is applied to actual projects.Compared with the previous manual adjustment based on experience,this method can give accurate coordinate values and make it clear that the steel bars need to be adjusted in size and direction with only one picture.Therefore,this method effectively improves the purpose and accuracy of rebar adjustment.Aiming at the problem of component pose measurement during the hoisting process,this paper proposes a three-dimensional pose monitoring system based on binocular vision.A coordinate system is established on site through the reconstructed three-dimensional point coordinates,so that the measured values and design values are converted into the same coordinate system for comparison.The feasibility of the visual measurement method is verified by numerical analysis of errors and indoor frame tests.Afterwards,it was verified by field tests that the system can effectively measure whether the components are well installed in place.Compared with the traditional method,the system can not only realize the rapid measurement of the spatial pose of components,but also reduce the investment of manpower and equipment.At the same time,in order to solve the problem of missed detection and false detection of targets in the construction site,this paper also proposes a target detection method,which has the advantages of fast detection rate and high accuracy.Through the application of the above methods in engineering,it can be seen that the construction process measurement technology based on machine vision can effectively improve the measurement efficiency of the construction site and save labor costs.It is foreseeable that machine vision measurement technology has bright prospects in the field of construction process measurement.