首页> 外文会议>IEEE International Conference on Robotics and Biomimetics >Research on 3D Surface Micro Topography Reconstruction and Mosaic from Multifocus Microscopic Image Sequences
【24h】

Research on 3D Surface Micro Topography Reconstruction and Mosaic from Multifocus Microscopic Image Sequences

机译:基于多焦点显微图像序列的3D表面微形貌重构和镶嵌研究

获取原文

摘要

Focus variation measurement (FVM), an optical contact-free method, has deep scientific research and application value for precision engineering, micro-manufacturing, quality inspection, biotechnology, clinical medicine and other fields. In order to reproduce a large-scale 3D surface micro topography and to realize the automatic measurement of the characteristic parameters of an object. The 3D model construction and mosaic (stitching) algorithm are excavated and studied based on 2D multifocus image sequence. In the research of 3D model reconstruction, a better focus measure is proposed based on the sum of the modified Laplacian (SML), which considers the relationship between the pixel and its diagonal direction pixels. And the 3D reconstruction models mosaic are guided by the mosaic of the 2D fused images for the reason that the pixels of the 2D fused image correspond to the depth points of the 3D reconstruction model. In order to verify the feasibility of the above methods, a comparative experiment is conducted on the standard triangular block model reconstruction with the SML and the proposed algorithm, which verifies the superiority of the proposed algorithm. Then, the two 3D models of the standard triangular block with overlapping parts are stitched and extracted the angle value. By comparing with the standard values, these angle values of 3D stitched model are within the standard value range, which verifies the feasibility of the stitching algorithm.
机译:聚焦变化测量(FVM)是一种无光学接触的方法,在精密工程,微制造,质量检查,生物技术,临床医学等领域具有深厚的科学研究和应用价值。为了再现大规模的3D表面微观形貌并实现对象特征参数的自动测量。挖掘和研究了基于2D多焦点图像序列的3D模型构建和镶嵌(拼接)算法。在3D模型重建的研究中,基于修正的拉普拉斯算子(SML)的总和,提出了一种更好的聚焦措施,该方法考虑了像素及其对角线方向像素之间的关系。由于2D融合图像的像素对应于3D重构模型的深度点,因此3D重构模型马赛克由2D融合图像的马赛克引导。为了验证上述方法的可行性,利用SML和所提出的算法对标准三角形块模型进行了比较实验,验证了所提出算法的优越性。然后,缝合具有重叠部分的标准三角形块的两个3D模型,并提取角度值。通过与标准值比较,这些3D缝合模型的角度值在标准值范围内,验证了缝合算法的可行性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号