Non-rigid registration of mural images and laser scanning data based on the optimization of the edges of interest
Non-rigid registration of mural images and laser scanning data based on the optimization of the edges of interest
复制标题
DOI：
10.1007/s11432-011-4440-3
复制
发表时间：
2012-04
期刊：
影响因子：
--
通讯作者：
Fan Zhang;Xianfeng Huang;Wei Fang;Deren Li
中科院分区：
文献类型：
--
作者：
Fan Zhang;Xianfeng Huang;Wei Fang;Deren Li
利用激光扫描数据对壁画图像进行配准定位和校正是古代壁画数字化保护的关键。提出了一种以激光回波强度信息为中介，基于感兴趣边缘优化的壁画图像与激光扫描数据的非刚性配准方法。首先，从激光回波强度信息生成强度图像，并使用刚性变换模型与壁画图像配准。其次，将壁画图像中的感兴趣边缘和强度图像的梯度场作为配准基元进行处理。最后，利用壁画图像和灰度图像刚性配准中的优化基，对每一条感兴趣的边缘进行配准。最后，利用优化后的边缘控制点建立两幅图像之间的非刚性变换模型，完成配准。实验结果表明，该方法对不同的数据集都能获得较高的配准精度。
The orientation and correction of mural images by registering them with laser scanning data is critical for the digital protection of ancient murals. This paper proposes a method for the non-rigid registration of mural images and laser scanning data based on the optimization of the edge of interest by using laser echo intensity information as an intermediary. First, the intensity image was generated from the laser echo intensity information, and registered with the mural image using a rigid transformation model. Second, the edges of interest in the mural image and the gradient field of the intensity images were processed as registration primitives. Third, every edge of interest was registered with the optimization base used in the rigid registration of the mural image and intensity image. Finally, the registration was completed after a non-rigid transformation model between these two images was constructed using the control points on the optimized edges. Our experimental results show that the proposed method can obtain high registration accuracy for different data sets.