Study On 3D Vision Measurement Based On Structured Light

With the development of economy,people's pursuit of quality of life is getting higher and higher,and the country pays more and more attention to the intelligent industry,and it is more and more important to embody computer vision.It can provide technologies such as virtual reality positioning in the field of reverse design of human body reconstruction medicine,and can reduce costs and improve efficiency.Structure light measurement is one of the important technology of computer vision,the structure light 3D measurement technology is a non-contact optical three-dimensional measurement method,its basic principle for the use of computer generated structure light pattern,using the projectors projected onto the object surface,then to decrypt pattern and a series of image processing to complete 3D reconstruction,however,most existing methods need to multiple projection design pattern to realize the closed form solution,which makes them unable to measure the dynamic object is proposed in this paper a kind of structured light 3D reconstruction method based on double cameras,and the effectiveness of the proposed method is verified by experiment.Based on the structure light,this paper mainly studies the 3D reconstruction method of dual cameras and applies the reconstruction method to the actual measurement.The main contents are asfollows.(1)In this paper,the theoretical methods and key technologies of three-dimensional measurement of structural light are studied in depth,the transformation relationship of four coordinate systems and internal and external parameters are analyzed from the mathematical point of views,and the principle of antipole geometry is also analyzed.(2)The hardware composition of the system built in this paper is stated,and the conversion relationship of the two cameras is analyzed.The dual camera calibration is completed according to the conversion relationship of the two cameras.The internal and external parameters of the camera are obtained.(3)Gathers the image of with a structured light pattern algorithm for image processing,this paper proposes a network segmentation method based on HSV,use this method to extract the pixel coordinates,using the presented RGB registration method of image can be filtered registration pixel coordinates,and use the similar triangle principle,imaging three-dimensional coordinates on the surface of the object.(4)Measured in practical application,this system to design good have to measure the size of the mold,measured in this paper,the structures,system error is 0.31 mm,the accuracy comparing with the device,this system is proved has better practicability and robustness of light according to the obtained 3 d on the surface of the heart and the object imaging surface obtained by the method of using analytic solution of coordinate point of three-dimensional coordinates,the use of cubic spline interpolation to complete 3 d reconstruction.