https://github.com/pei-mao/cuda-erodeanddilate

This is a small implementation example of image processing using CUDA technology, demonstrating basic operation methods.
https://github.com/pei-mao/cuda-erodeanddilate

cuda-programming image-processing

Last synced: 11 months ago
JSON representation

This is a small implementation example of image processing using CUDA technology, demonstrating basic operation methods.

• Host: GitHub
• URL: https://github.com/pei-mao/cuda-erodeanddilate
• Owner: Pei-mao
• Created: 2024-07-10T04:44:59.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2024-07-18T02:55:49.000Z (almost 2 years ago)
• Last Synced: 2025-01-02T23:19:37.158Z (over 1 year ago)
• Topics: cuda-programming, image-processing
• Language: C++
• Homepage:
• Size: 639 KB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Cuda-ErodeAndDilate

A small implementation of image processing operations (erosion and dilation) using CUDA for parallel computation.

| Raw Image | Erosion Image | Dilation Image |

|:---------:|:-------------:|:--------------:|

| ![Raw Image](images/finger.jpg) | ![Erosion Image](images/Erosion.jpg) | ![Dilation Image](images/Dilation.jpg) |

## Process

1. **Read the image**: Read the image from the specified path and convert it to grayscale.

2. **Allocate memory**: Allocate memory on the GPU to store the input and output image data.

3. **Set CUDA kernel parameters**: Define the number of threads per block and calculate the number of blocks needed.

4. **Execute CUDA kernel**: Execute the CUDA kernels for erosion and dilation operations.

5. **Retrieve results**: Copy the computed results from the GPU back to the CPU and save the resulting images to the specified path.

6. **Perform operations using OpenCV**: Perform erosion and dilation operations using OpenCV's `erode` and `dilate` functions, and save the results.