# Parallel image processing

**URL:** <https://forums.developer.nvidia.com/t/parallel-image-processing/19780>\
**Category:** CUDA Programming and Performance\
**Created:** [November 16, 2010, 5:50pm UTC](https://forums.developer.nvidia.com/t/parallel-image-processing/19780 "2010-11-16T17:50:52Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![JackyLikesGpu](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@JackyLikesGpu](https://forums.developer.nvidia.com/u/JackyLikesGpu)\
**Post date:** [November 16, 2010, 5:50pm UTC](https://forums.developer.nvidia.com/t/parallel-image-processing/19780/1 "2010-11-16T17:50:52Z")

</div>

Hi All,

I am having a Geforce GTX 460 and am planning to do some parallel image processing.  
Problem def: I have an image A. I want to apply an image processing algo (for example lets say image dilation) for different regions of the image (the shape of the ROI is not fixed). How can i do this in parallel such that each ROI of the image is processed by different cuda blocks and hence speed up the process.

Thanks in advance for all the help.

Jacky,

---

<div class="post-metadata">

**Author:** ![JackyLikesGpu](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@JackyLikesGpu](https://forums.developer.nvidia.com/u/JackyLikesGpu)\
**Post date:** [November 16, 2010, 5:50pm UTC](https://forums.developer.nvidia.com/t/parallel-image-processing/19780/2 "2010-11-16T17:50:52Z")

</div>

Hi All,

I am having a Geforce GTX 460 and am planning to do some parallel image processing.  
Problem def: I have an image A. I want to apply an image processing algo (for example lets say image dilation) for different regions of the image (the shape of the ROI is not fixed). How can i do this in parallel such that each ROI of the image is processed by different cuda blocks and hence speed up the process.

Thanks in advance for all the help.

Jacky,

---

<div class="post-metadata">

**Author:** ![Dittoaway](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@Dittoaway](https://forums.developer.nvidia.com/u/Dittoaway)\
**Post date:** [November 16, 2010, 7:06pm UTC](https://forums.developer.nvidia.com/t/parallel-image-processing/19780/3 "2010-11-16T19:06:22Z")

</div>

Tell us how you will do it serially and we’ll help you do it in parallel.

---

<div class="post-metadata">

**Author:** ![Dittoaway](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@Dittoaway](https://forums.developer.nvidia.com/u/Dittoaway)\
**Post date:** [November 16, 2010, 7:06pm UTC](https://forums.developer.nvidia.com/t/parallel-image-processing/19780/4 "2010-11-16T19:06:22Z")

</div>

Tell us how you will do it serially and we’ll help you do it in parallel.

---

<div class="post-metadata">

**Author:** ![JackyLikesGpu](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@JackyLikesGpu](https://forums.developer.nvidia.com/u/JackyLikesGpu)\
**Post date:** [November 16, 2010, 9:56pm UTC](https://forums.developer.nvidia.com/t/parallel-image-processing/19780/5 "2010-11-16T21:56:25Z")

</div>

> [@](#):
>
> Tell us how you will do it serially and we’ll help you do it in parallel.

Thanks for a quick one.

I have a label for each pixel in the image (say 1 to 10)

I store the pixels corresponding to same label as sub images (SI) (here number of SIs = 10)

serial pseudo code

for i=1 to 10

SI\_dilated= perform image dilation on SI(i);

get some value from SI\_dilated;

end for loop

I have large number of SIs and hence want to make use of parallelization.

---

<div class="post-metadata">

**Author:** ![JackyLikesGpu](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@JackyLikesGpu](https://forums.developer.nvidia.com/u/JackyLikesGpu)\
**Post date:** [November 16, 2010, 9:56pm UTC](https://forums.developer.nvidia.com/t/parallel-image-processing/19780/6 "2010-11-16T21:56:25Z")

</div>

> [@](#):
>
> Tell us how you will do it serially and we’ll help you do it in parallel.

Thanks for a quick one.

I have a label for each pixel in the image (say 1 to 10)

I store the pixels corresponding to same label as sub images (SI) (here number of SIs = 10)

serial pseudo code

for i=1 to 10

SI\_dilated= perform image dilation on SI(i);

get some value from SI\_dilated;

end for loop

I have large number of SIs and hence want to make use of parallelization.