This ReadMe provides a comprehensive guide on running the CUDA-accelerated image processing code, including Gaussian blur, erosion, dilation, and unsharp mask algorithms. The code is designed to enhance image processing performance using parallel computing techniques with CUDA. Below are the step-by-step instructions for running the code.
Before running the code, ensure that the following prerequisites are met:
- NVIDIA GPU with CUDA support.
- CUDA Toolkit installed on the system.
- C++ compiler that supports CUDA.
Clone the repository containing the CUDA-accelerated image processing code to your local machine.
git clone <https://github.com/omarkhaled2001/PC_Big_Assignment.git>
Navigate to the directory ".\Colab" where the code is located and open Colab and choose file -> open notebook then choose file PC.ipynb, then upload all other files in same directory .
Once the code is successfully Uploaded, choose RunTime -> RunTimeType -> T4 GPU then run all cells.
- The code includes implementations of Gaussian blur, erosion, dilation, and unsharp mask algorithms using CUDA.
- Performance metrics and comparison charts are generated to illustrate the improvements achieved by CUDA-accelerated image processing.
- The code is optimized for parallelism and efficient memory access to maximize performance.
The CUDA-accelerated image processing algorithms have various applications, including:
- Medical Imaging
- Computer Vision
- Robotics
- Remote Sensing
- Document Processing
This ReadMe provides a guide for running the CUDA-accelerated image processing code, offering enhanced performance and potential applications in various fields. For further details on the code implementation and optimization, refer to the source code and associated documentation.
Thank you for using the CUDA-accelerated image processing code. If you have any questions or encounter issues, please refer to the documentation or contact the repository owner for assistance.