此为 Mathematica 8 文档,内容基于更早版本的 Wolfram 语言
查看最新文档(版本11.1)

CUDAOpening


gives the morphological opening of img with respect to a range-r square.

gives the morphological opening of list with respect to a range-r square.

gives the morphological opening of mem with respect to a range-r square.
  • The CUDALink application must be loaded using Needs.
  • works with binary, grayscale, and other images.
  • If multiple image channels are present, operates on each of them separately.
  • The following options can be given:
"Device"Automaticdevice used for CUDA computation
Padding"Fixed"value used on boundary
"OutputMemory"NoneCUDAMemory where output is stored
First, load the CUDALink application:
Use morphological opening to extract objects larger than the structuring element:
The radius can be varied:
works with lists:
works with graphics objects:
Using CUDAMemory, you can speed up computation by reducing the time spent on memory operations. This loads memory onto the GPU:
This performs the opening operation:
This gets the memory back to the CPU:
Memory must be deleted:
First, load the CUDALink application:
In[1]:=
Click for copyable input
Use morphological opening to extract objects larger than the structuring element:
In[2]:=
Click for copyable input
Out[2]=
 
The radius can be varied:
In[1]:=
Click for copyable input
Out[1]=
 
works with lists:
In[1]:=
Click for copyable input
Out[1]//MatrixForm=
 
works with graphics objects:
In[1]:=
Click for copyable input
Out[1]=
 
Using CUDAMemory, you can speed up computation by reducing the time spent on memory operations. This loads memory onto the GPU:
In[1]:=
Click for copyable input
This performs the opening operation:
In[2]:=
Click for copyable input
Out[2]=
This gets the memory back to the CPU:
In[3]:=
Click for copyable input
Out[3]=
Memory must be deleted:
In[4]:=
Click for copyable input
Out[4]=