Rock Image C++ is simple GUI application build in C++ and Qt to currentLayer layers from rocks for neural networks training.
Rock Image C++ is simple GUI application build in C++ and Qt to currentLayer layers from rocks for neural networks training.
- Check for ImageMagik Lib to convert any image type to PNG or other file format that Qt is able to work with.
- Create script to convert images.
- Customize pen color and size for each layer.
- Fix Zoom In and Out.
- Add AI services to provide image segmentation.
- Provide classes and functions to create and run AI models.
- Provide classes and functions to create a dataset.
- Create Docker image
First create a folder to receive the build files.
$ mkdir build
Then, with the cmake
run the follow command in your terminal.
$ cmake -B build -S .
Now, inside the build folder run the make command to compile the project.
$ cd build && make
After the compilation is completed run the executable.
$ ./rock_image_cpp
So far this project only support few image formats, such as JPG, BMP and PNG. Use the convimg
script to convert the images from other formats to PNG and use with the project.
# tools/convimg <director-with-images> <original-format>
$ tools/convimg ~/Images tif
It's possible to run this project on a docker container. First make sure that docker and docker compose are installed locally. If not follow the instructions in the official page.
$ docker -v
$ docker-compose -v
Then grant access to everyone with xhost
. It's going to be necessary to run the GUI.
$ xhost +
Build the image and then use the tools/rundocker
script with the images repository as a parameter. It's necessary because the docker container will run isolated from the rest of the machine, therefore a volume (a folder) must to be shared.
$ docker build -t rockimage .
$ tools/rundocker "/path/to/images"