基于网络节点相似性的髋关节点云配准算法  

Hip Joint Point Cloud Registration Algorithm Based on Network Node Similarity

在线阅读下载全文

作  者:邓正强 徐琦[1] 聂雨晨 周迪斌[1] DENG Zhengqiang;XU Qi;NIE Yuchen;ZHOU Dibin(School of Information Science and Technology,Hangzhou Normal University,Hangzhou 311121,China)

机构地区:[1]杭州师范大学信息科学与技术学院,浙江杭州311121

出  处:《杭州师范大学学报(自然科学版)》2023年第2期135-140,172,共7页Journal of Hangzhou Normal University(Natural Science Edition)

基  金:国家自然科学基金联合基金重点项目(U21A20466).

摘  要:针对基于聚类和距离的点云去除异常值算法在髋关节点云上容易去除正常值,导致点云结构发生改变等问题,本文提出一种随机抽样一致性(random sample consensus,RANSAC)拟合圆算法与包围球去除离群值算法,利用RANSAC算法提出距离点云主体较远的异常值,使用包围球算法去除髋臼窝中悬浮的异常值.同时,针对迭代最邻近点算法(iterative closest point,ICP)在选取初始点对时偏离真实值较大,容易导致迭代次数过多,陷入局部最优解的情况,本文提出一种基于网络节点相似性筛选初始点对方法,通过选取最接近真实值的点对进行迭代优化.结果表明,改进算法能够有效剔除髋关节表面点云中的异常值,选取准确的初始点对,实现任意位姿下两组点云的快速、精确配准.The point cloud outlier removal algorithm based on clustering and distance can easily remove the normal value on the hip joint point cloud,resulting in the change of the point cloud structure.A random sample consensus(RANSAC)fitting circle algorithm and an outlier removal algorithm with bounding spheres were proposed in this study.Using RANSAC algorithm,the outliers far from the main body of the point cloud were suggested,and the outliers suspended in the acetabular fossa were removed by the bounding spheres algorithm.Meanwhile,the iterative closest point(ICP)algorithm deviated greatly from the true value when selecting the initial point pair,which easily resulted in too many iterations and fell into the local optimal solution.Therefore,a method of filtering the initial point pair based on the similarity of network nodes was proposed,and the point closest to the real value was selected for iterative optimization.The results showed that the improved algorithm could effectively remove the outliers in the point cloud of the hip joint surface,effectively selected the exact initial point pair,and achieved fast and accurate registration of two groups of point clouds in any position.

关 键 词:点云配准 包围球 ICP算法 网络节点相似性 

分 类 号:TP317.4[自动化与计算机技术—计算机软件与理论]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象