CUDADilation
CUDADilation[img,r]
gives the morphological dilation of img with respect to a range-r square.
CUDADilation[mem,r]
gives the morphological dilation of list with respect to a range-r square.
CUDADilation[mem,r]
gives the morphological dilation of mem with respect to a range-r square.
Details and Options
- The CUDALink application must be loaded using Needs["CUDALink`"].
- CUDADilation works with binary, grayscale, and other images.
- CUDADilation works with CUDAMemory and lists.
- CUDADilation[image,r] is equivalent to Dilation[image,r].
- If multiple image channels are present, CUDADilation 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
Examples
Basic Examples (6)
First, load the CUDALink application:
This removes smaller, dark features:
Dilation can be used to perform a min filter on an input list:
CUDADilation works with graphics objects:
Using CUDAMemory, one can speed up computation by reducing the amount of memory operations. This loads memory onto the CPU:
This performs the dilation operation:
Text
Wolfram Research (2010), CUDADilation, Wolfram Language function, https://reference.wolfram.com/language/CUDALink/ref/CUDADilation.html.
CMS
Wolfram Language. 2010. "CUDADilation." Wolfram Language & System Documentation Center. Wolfram Research. https://reference.wolfram.com/language/CUDALink/ref/CUDADilation.html.
APA
Wolfram Language. (2010). CUDADilation. Wolfram Language & System Documentation Center. Retrieved from https://reference.wolfram.com/language/CUDALink/ref/CUDADilation.html