Important
The ucasML program is designed to run on a Linux operating system.
The OpenCV .so libraries included in opencv_libs.zip have been compiled with a dependency on GLIBCXX_3.4.32. This version of GLIBCXXis part of libstdc++, which comes with GCC (GNU Compiler Collection) version 12 or later.
To ensure compatibility, you need to have GCC version 12 or later installed on your system. You can check your current GCC version by running the following command:
gcc --versionCaution
Upgrading your GCC is strongly discouraged: such an action may lead to a broken system.
If the above requirement is not met, install or upgrade to a distro which supports GCC version 12 (or later), like the ones we used: see Tested Linux Distributions.
Note
This program has been installed on these Linux distros.
Tip
Click the badges to visit the distro official website.
Follow the steps below to ensure all dependencies are met and the program runs successfully.
Note
This guide will help you install and run the ucasML program on Ubuntu.
Warning
Despite all the following commands have been already tested, everything you do is "at your own risk".
➡️ Clone the Repository ⬅️
First, clone the repository from GitHub:
git clone https://github.com/giusalfieri/IPA_Project.git
cd IPA_Project➡️ Making the ucasML file executable ⬅️
Navigate to the bin directory inside the ucasML_package directory and make the ucasML file executable:
cd ucasML_package/bin
chmod +x ucasML➡️ Unzip OpenCV Precompiled .so Libraries ⬅️
Extract the content of opencv_libs.zip into the opencv_libs folder.
Tip
After the extraction, you can delete opencv_libs.zip.
➡️ Set the Library Path Permanently ⬅️
To make the library path setting permanent, you need to add the library path to your .bashrc file (.zshrc for zsh shell, or the config file of the shell you are using):
nano ~/.bashrcAdd the following line at the end of the file:
export LD_LIBRARY_PATH=/path/to/ucasML_package/opencv_libs:$LD_LIBRARY_PATHImportant
Replace /path/to/ucasML_package/opencv_libs with the actual path to the opencv_libs directory on your system. For example, if you have extracted the package to your home directory, the line above would look like this:
export LD_LIBRARY_PATH=/home/your-username/ucasML_package/opencv_libs:$LD_LIBRARY_PATHSave the file and exit the text editor. Then, apply the changes:
source ~/.bashrc➡️ Check and Install Missing Dependencies ⬅️
Use ldd to check for any missing dependencies:
ldd ./ucasMLInstall any missing libraries as indicated by the ldd output. For example, if a library is missing, you can typically install it by installing the corresponding package using apt (or the package manager of the distro being used):
sudo apt install <missing-library>Note
Additionally, it may be necessary to create symbolic links to resolve dependencies required by the application. For example, on Ubuntu 24.04 and Kali, we needed to create the following symbolic link:
sudo ln -s /usr/lib/x86_64-linux-gnu/libdc1394.so.25 /usr/lib/x86_64-linux-gnu/libdc1394Note
libjpeg8 is required by the application, but it is obsolete and has been substituted by libjpegturbo packages. Some changes may be necessary.
If any problems persist, consider removing conflicting packages and installing libjpeg8.
Once all dependencies are resolved, you can run the ucasML program from the bin directory:
This program was developed by Alessandro Bria, currently Associate Professor at the University of Cassino.