Three-dimensional Reconstruction Of Object Surface Based On Focusing Method

Vision-based 3D reconstruction technology is widely used in modern society,including industrial detection,medical systems,face recognition,road traffic,and machine vision.It has a very broad application prospect,and is also a current Popular topics for research.With the continuous development of computer vision technology,3D reconstruction measurement technology also plays a role in more fields.In industrial inspection,the three-dimensional reconstruction of objects using three-dimensional methods is based on the defect detection or size measurement of objects based on machine vision,so as to achieve the detection of the appearance quality of the target object.This thesis is devoted to the research on the three-dimensional reconstruction of the surface of small components based on the automatic focusing method,focusing on the depth of focus(DFF)method in the focusing method,and reconstructing the three-dimensional shape of the object surface from the focused image sequence It uses a telecentric lens to capture image sequences with the same resolution and different depth of focus to reconstruct the three-dimensional shape of the object.On the basis of the existing equipment in the laboratory,the surface of the object can be reconstructed by digital image analysis processing without increasing the hardware cost The three-dimensional shape has the advantages of non-contact,simple hardware equipment and realization of the restoration of the shape of small objects.This article is divided into six chapters in total.The main content is to discuss the research status,existing deficiencies,algorithm improvements and the construction and implementation of software and hardware equipment measurement systems based on the focus adjustment method of 3D reconstruction.The software c++ builder programming tool is used to realize the functions of electric slide control and image acquisition,and the algorithm compilation software halcon is used to realize the function of recovering the three-dimensional shape of the object surface.The completed work includes the following aspects:1 According to the actual requirements of the project,a fully automatic hardwareslide system was designed.The functions of mobile slide control and image data acquisition were realized through programming software.Finally,the image acquisition and 3D reconstruction were combined to realize the overall software and hardware system.Combine and display on the same interface.2 This thesis uses the halcon algorithm programming software to realize the 3D reconstruction of the DFF algorithm.It restores the depth information of the object from a series of pictures in different focus states,restores the 3D topography of the object surface,and realizes 3D reconstruction.The halcon algorithm is imported into the c ++ builder programming tool and the image acquisition algorithm is combined into a complete program,which finally displays the results of image acquisition and3 D reconstruction on the interface.3 The commonly used focus evaluation algorithms are compared,the aggregation evaluation algorithms are compared,the deficiencies are analyzed,and choose the focus evaluation algorithm suitable for this topic.Through the built experimental system,the image sequence of unsynchronized length is collected,and the error analysis of the obtained all-focus image is performed.It is verified that the smaller the step size,the better the image quality.Obtain point cloud data from the obtained all-focus image,fit the outer circle diameter according to the point cloud data,and compare and analyze the obtained diameter with the actual diameter of the sample.The absolute errors of the outer diameter of the outer diameter under the unsynchronized length are all less than 0.03.