| Laser fusion is an important research area of controllable fusion and an important direction of overcoming energy crisis .In the experiment of laser fusion , it's necessary to use targets that has high quality and accuracy.So the production, assembly and measurement of the targets are important steps in laser fusion experiment.This essay researches on how to measure the position of the parts of the targets when assembling to monitor the process of the assembly ,ensuring that the targets can be assembled accurately.The target is very small,and has low hardness,and is deformed easily.So this essay chooses the measurement based on machine vision.The basic measurement process is described later.First,we get the target's images.Then we extract target's information from the images.And then we transform the information from images to space.At last, we can get the target's parameters in the space.This essay designs proper image processing algorithms,first median-filtering, then Canny-algorithm ,enabling us to get the target's accurate and complete edge with low noise. This essay deeply researches the detection of lines,circles and ellipses in the iamge .Especially the methods for the detection of circles and ellipses have more improvement and new theories than the essays before,reducing much calculation and time .To calibrate the cameras,this essay improves Tsai's method ,calibrating the image center and the scale factor in advance and doing nonlinear optimization for them with other parameters.For three-dimensional reconstruction, this essay gives new reconstruction method for points and lines . Especially the overall reconstruction method for lines reduces much calculation and time . At last, this essay designs a set of methods to measure the error of the position of the the middle of the target ball and the error of the observation window.After experimental verification ,we know that the errors of the measurement results are all under 5μm , proving that this essay's research has good effect. |
PDF Full Text Request