https://github.com/notkartikye/cuda-image-box-filters

🖼️ CUDA-powered tool for applying box filters to a large amount of images
https://github.com/notkartikye/cuda-image-box-filters

cuda cuda-library cuda-programming npp

Last synced: about 2 months ago
JSON representation

🖼️ CUDA-powered tool for applying box filters to a large amount of images

• Host: GitHub
• URL: https://github.com/notkartikye/cuda-image-box-filters
• Owner: notkartikye
• Created: 2024-07-01T09:15:27.000Z (8 months ago)
• Default Branch: main
• Last Pushed: 2024-07-15T20:37:47.000Z (7 months ago)
• Last Synced: 2024-11-06T08:18:32.766Z (3 months ago)
• Topics: cuda, cuda-library, cuda-programming, npp
• Language: C++
• Homepage:
• Size: 4.08 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# CUDA at Scale Independent Project

## Description

This program applies CUDA box filters to all images in a folder, using NPP's `nppiFilterBoxBorder_8u_C1R` routine.

(Based on nvidia's sample code: https://github.com/NVIDIA/cuda-samples/tree/master/Samples/4_CUDA_Libraries/boxFilterNPP)

## Implementation Details

`boxFilterNPP.cpp` is the main implementation, it loops through all files in the input folder, reads the image,

copies to device, applies the filter, and writes the result to the output folder.

`Common` contains the helper code for error checking and image file handling.

`data` contains example input images.

`output` contains output for the example inputs.

`Makefile` for compiling the code.

`run.sh` contains scripts for building the code and running on example inputs.

## Running the Program

Run the following command, with the `-input` flag to specify the input folder to use:

  ```bash

  run.sh

  ```