| With the rapid development of computer vision,three-dimensional(3D)identification technology has been widely used in the fields of industrial bulk sorting and robot grasping.The 3D identification technology makes up for the shortcomings of two-dimensional identification technology.It not only improves the accuracy of bulk identification,but also expands the scope of visual measurement.The 3D identification technology can achieve3 D non-contact measurement of industrial material.It is applied to online detection and identification of industrial material in production line.In this paper,a 3D identification system of bulk material based on line-structured light is designed.The feasibility of this system is validated by the experiment of identificating the industrial bulk material.In this thesis,the related technology of line-structured light 3D recognition is studied firstly.The overall scheme of 3D identification system is proposed.The hardware and software design of the system is completed.The calibration method of identification system is studied secondly.Based on the principle of calibration,a non-linear camera imaging model is established to achieve efficient calibration of camera.The simple and efficient calibration of light plane is attained when adopting the extraction method of feature points in light plane.Then the improved Steger algorithm is applied to extract the centerline of light stripe,which improves the accuracy and efficiency of extraction.Finally,the center of light stripe extracted by each frame is spliced.The 3D reconstruction is completed,and the 3D contour of the measured object is identified.Theoretical analysis and experimental results show that the proposed 3D identification system can effectively realize the 3D identification of industrial material and improve the accuracy of 3D recognition.Therefore,the practical application of 3D recognition technology has important theoretical significance and engineering value. |
PDF Full Text Request