Image_processing_tool

A multi-threaded image processing application that does sharpening and edge detection on a PNG using highpass and Sobel filtering. Can be run using 1,2,4,8,16,32 threads for more efficient processing. It creates a copy of the original with the filters applied and DOES NOT alter the original image.

Features

• Sharpen: Sharpen outlines in a PNG.
• Edge detect: Finds the boundaries of a PNG.
• Multi-Threaded: Use more threads for faster results (depending on your CPU).

Installation

1. Clone the repository:

   git clone https://github.com/marwanthestudent/Image_processing_tool.git

2. Install dependencies: Ensure you have C installed with the required libraries (libpng, zlib):
   For more info -> http://www.libpng.org/pub/png/libpng.html
   If you have a package manager like homebrew or pacman (whatever your OS and manager is)

   homebrew install libpng

3. Compilation path: Ensure you have the directory libpng16 in your include path so the header file can be recognized (include <png.h>).

    -I<whatever location the include directory is in>

   Check Makefile for more comments.

4. Linker path: Ensure that the linker has the path to require files

    -L<whatever location libpng is in> -lpng16 -lz

   Check Makefile for more comments.

Usage

1. Running the executable:

   ./image_process <relative path of image> <relative path and name desired of output> <edge, sharpen> <No of threads: 1,2,4,8,16,32 >

Example Output

An example of how the output would look using the example imgs folder:

Original PNG	Output PNG after edge detection

Original PNG	Output PNG after sharpening

Additional Notes

There is a TROFF manual I made in the manual directory so the program can feel more like a command (mainly for learning purposes)

Screenshot from terminal

To display it, move image_process.1 to your manual's folder (in super user), for me it was

sudo mv image_process.1 /usr/Local/share/man/man1

Then from the terminal you can display the manual by simply

man Image_process

Future Work

• Implement Gaussian blurring and Unsharp. I had trouble debugging so I did not include this feature in the current version.

License

This project is licensed under the MIT License. See the LICENSE file for details.