CUDAImageDivide
This feature is not supported on the Wolfram Cloud.

CUDAImageDivide[img,x]
divides each channel value in img by an amount x.

CUDAImageDivide[mem,x]
divides each channel value in mem by an amount x.

CUDAImageDivide[img1,img2]
gives an image in which each pixel is the division of the corresponding pixels in img1 and img2.

CUDAImageDivide[mem1,mem2]
gives a CUDAMemory in which each pixel is the division of the corresponding pixels in mem1 and mem2.

更多信息更多信息

  • The CUDALink application must be loaded using Needs["CUDALink`"].
  • CUDAImageDivide works with binary, grayscale, and other images.
  • CUDAImageDivide works with CUDAMemory and lists.
  • CUDAImageDivide[image1,x] is equivalent to CUDAImageMultiply[image1,1/x].
  • The image returned by CUDAImageDivide[image,] has the same dimensions as image.
  • image1 and image2 must be the same dimensions in CUDAImageDivide[image1,image2].
  • If image2 is a single-channel image, each channel of image1 is divided by image2 in CUDAImageDivide[image1,image2].
  • In CUDAImageDivide[image,x], the number x is normally in the range 0 to 1 and is not defined at 0.
  • CUDAImageDivide[image,x] gives an image with the same underlying data type as image, clipping or truncating values if necessary.
  • CUDAImageDivide[image1,image2] gives an image with the type of image1, clipping or truncating values if necessary.
  • The following options can be given:
  • "Device"Automaticdevice used for CUDA computation
    "OutputMemory"NoneCUDAMemory where output is stored

范例范例打开所有单元关闭所有单元

基本范例  (3)基本范例  (3)

First, load the CUDALink application:

In[1]:=
Click for copyable input

This divides each pixel of the input image by a constant value 0.5:

In[2]:=
Click for copyable input
Out[2]=

This divides each pixel of the input image by a constant value 2.0:

In[1]:=
Click for copyable input
Out[1]=

CUDAImageDivide can be used to multiply rank 2 or 3 lists:

In[1]:=
Click for copyable input
Out[1]//MatrixForm=