CUDAOpening[img,r]
gives the morphological opening of img with respect to a range-r square.
CUDAOpening[list,r]
gives the morphological opening of list with respect to a range-r square.
CUDAOpening[mem,r]
gives the morphological opening of mem with respect to a range-r square.
CUDAOpening
CUDAOpening[img,r]
gives the morphological opening of img with respect to a range-r square.
CUDAOpening[list,r]
gives the morphological opening of list with respect to a range-r square.
CUDAOpening[mem,r]
gives the morphological opening of mem with respect to a range-r square.
更多信息和选项
- The CUDALink application must be loaded using Needs["CUDALink`"].
- CUDAOpening works with binary, grayscale, and other images.
- CUDAOpening[image,r] is equivalent to Opening[image,r].
- If multiple image channels are present, CUDAOpening operates on each of them separately.
- The following options can be given:
-
"Device" Automatic device used for CUDA computation Padding "Fixed" value used on boundary "OutputMemory" None CUDAMemory where output is stored
范例
基本范例 (5)
First, load the CUDALink application:
Needs["CUDALink`"]Use morphological opening to extract objects larger than the structuring element:
CUDAOpening[[image], 4]CUDAOpening[[image], 7]MatrixForm[CUDAOpening[RandomInteger[{0, 255}, {5, 5, 3}], 1]]CUDAOpening works with graphics objects:
CUDAOpening[\!\(\*Graphics3DBox[«7»]\), 5]Using CUDAMemory, you can speed up computation by reducing the time spent on memory operations. This loads memory onto the GPU:
mem = CUDAMemoryLoad[[image]];This performs the opening operation:
CUDAOpening[mem, 1]This gets the memory back to the CPU:
CUDAMemoryGet[%]CUDAMemoryUnload[mem]相关指南
-
▪
- CUDALink
文本
Wolfram Research (2010),CUDAOpening,Wolfram 语言函数,https://reference.wolfram.com/language/CUDALink/ref/CUDAOpening.html.
CMS
Wolfram 语言. 2010. "CUDAOpening." Wolfram 语言与系统参考资料中心. Wolfram Research. https://reference.wolfram.com/language/CUDALink/ref/CUDAOpening.html.
APA
Wolfram 语言. (2010). CUDAOpening. Wolfram 语言与系统参考资料中心. 追溯自 https://reference.wolfram.com/language/CUDALink/ref/CUDAOpening.html 年
BibTeX
@misc{reference.wolfram_2026_cudaopening, author="Wolfram Research", title="{CUDAOpening}", year="2010", howpublished="\url{https://reference.wolfram.com/language/CUDALink/ref/CUDAOpening.html}", note=[Accessed: 15-August-2026]}
BibLaTeX
@online{reference.wolfram_2026_cudaopening, organization={Wolfram Research}, title={CUDAOpening}, year={2010}, url={https://reference.wolfram.com/language/CUDALink/ref/CUDAOpening.html}, note=[Accessed: 15-August-2026]}