基于IGA-SVR随机Hough变换的预制构件螺栓孔节点定位

Bolt Hole Nodes Location of Prefabricated Components Based on IGA-SVR Combined with Random Hough Transform

摘要:

针对预制构件螺栓孔节点定位时由于工业环境下光照不均匀且螺栓孔并非规则平滑圆形，导致定位精度不高的问题；提出一种引入改进遗传算法优化支持向量回归分析（IGA-SVR）的随机Hough变换圆形定位算法；对遗传算法引入收缩包围与螺旋更新机制，提升算法局部搜索能力；同时改进交叉变异算子，引入收敛因子克服遗传算法后期收敛速度慢的问题。利用改进的遗传算法对支持向量回归模型进行参数寻优，通过圆形训练样本训练出逼近螺栓孔的超平面方程；采用此模型上的三点进行随机Hough变换定位圆，并利用双目视觉算法得到圆形三维坐标；通过四个标准测试函数及混凝土预制构件模型螺栓孔节点定位实验验证所提算法的有效性，结果表明，改进的算法优化性能更佳，在结合随机Hough变换定位圆时，明显提高了定位精度，满足工程测量要求。

Abstract:

For prefabricated components bolt hole nodes positioning, due to uneven lighting in the industrial environment and bolt holes are not regular smooth round, resulting in poor positioning accuracy.The paper proposed a random Hough transform circular localization algorithm introducing an improved genetic algorithm to optimize support vector regression analysis (IGA-SVR).The genetic algorithm is improved by introducing a shrinkage bracket and spiral update mechanism to enhance the local search ability of the algorithm; at the same time, the cross-variance operator is improved and a convergence factor is introduced to overcome the problem of slow convergence of the genetic algorithm at the later stage. The improved genetic algorithm is used to optimize the parameters of the support vector regression model, and the hyperplane equation approximating the bolt hole is trained by circular training samples.The three points on this model are used to locate the circle by random Hough transform, and the three-dimensional coordinates of the circle are obtained by binocular vision algorithm.The effectiveness of the proposed algorithm is verified by four standard test functions and bolt hole node positioning experiments of concrete precast component model. The results show that the improved algorithm has better optimization performance. When combined with random Hough transform positioning circle, the positioning accuracy is significantly improved and meets the requirements of engineering measurement.