首页> 外文会议>Conference on vision geometry >Connectivity-preserving parallel operators in 2D and 3D images
【24h】

Connectivity-preserving parallel operators in 2D and 3D images

机译:在2D和3D图像中保留连接性的并行运算符

获取原文

摘要

Abstract: ivity preservation is a concern in the design of parallel reduction processes for 2D and 3D image processing algorithms. Algorithm designers need efficient and available connectivity preservation tasks to prove algorithm correctness. Although efficient 2D tests are known, efficient 3D tests still need to be developed. We review earlier results for 2D connectivity preservation tests and demonstrate several 'design spaces' for classes of parallel reduction operators including subiteration and subfields approaches. We then extend certain 'path based' tests from the 2D to the 3D case and show efficient realizations for all but one test for fully parallel reduction operators. Very efficient tests are determined for 3D subfields reduction operators.!36
机译:摘要:在2D和3D图像处理算法的并行归约过程设计中,效率保持是一个需要关注的问题。算法设计人员需要有效且可用的连接保留任务来证明算法的正确性。尽管有效的2D测试是已知的,但仍然需要开发有效的3D测试。我们回顾了2D连通性保留测试的早期结果,并为并行归约算子类别(包括子迭代法和子域方法)展示了几个“设计空间”。然后,我们将某些“基于路径”的测试从2D扩展到3D情况,并展示了针对除完全并行归约算子以外的所有测试的有效实现。为3D子场归约算子确定了非常有效的测试!36

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