sablib

A C++ library for signal smoothing and baseline estimation, powered by Eigen.

Features

Baseline Estimation

• Linear/Polynomial: Robust linear or polynomial baseline fitting to anchor points.
• Spline: Cubic Spline interpolation-based baseline estimation.
• SMA: Iterative Simple Moving Average for baseline estimation.
• SNIP: Statistics-sensitive Non-linear Iterative Peak-clipping.
• ModPoly: Modified Polynomial.
• IModPoly: Improved ModPoly.
• Backcor: Iterative polynomial fitting with a non-quadratic cost function.
• Goldindec: Goldindec algorithm (a variation of Backcor).
• AsLS: Asymmetric Least Squares.
• airPLS: Adaptive Iteratively Reweighted Penalized Least Squares.
• arPLS: Asymmetrically Reweighted Penalized Least Squares.
• psalsa: Peaked Signal's Asymmetric Least Squares Algorithm.
• BEADS: Baseline Estimation And Denoising using Sparsity.

Smoothing

• Moving Average: Standard simple moving average filter.
• Weighted Moving Average: Configurable weighted moving average filter.
• Moving Median: Moving median.
• Savitzky-Golay: Savitzky-Golay filter for simultaneous smoothing and differentiation.
• Whittaker: Whittaker smoother based on Penalized Least Squares.
• P-Spline: Penalized B-Spline smoothing.

Requirements

• C++ Standard: C++17 or later.
• Dependencies: Eigen 3.4.0 or later.

Installation & Build

Using vcpkg

sablib provides a vcpkg port. First, clone this repository:

git clone https://github.com/Izadori/sablib

Then install with the --overlay-ports option:

vcpkg install sablib --overlay-ports=<path/to/sablib>/vcpkg/ports

In your CMakeLists.txt:

find_package(sablib CONFIG REQUIRED)
target_link_libraries(my_app PRIVATE sablib::sablib)

Using Conan

sablib provides a Conan recipe. First, clone this repository:

git clone https://github.com/Izadori/sablib

Then install and build:

cd sablib
conan install conan/sablib --output-folder=conan/build --build=missing
conan build conan/sablib --output-folder=conan/build

In your CMakeLists.txt:

find_package(sablib CONFIG REQUIRED)
target_link_libraries(my_app PRIVATE sablib::sablib)

Building manually

sablib can also be built as a static library using CMake directly:

mkdir build
cd build
cmake ..
cmake --build .

If Eigen3 is not found, please try cmake .. -DCMAKE_PREFIX_PATH="path/to/Eigen3".

Usage

Integrating with CMake

Via vcpkg

After installing with vcpkg, use find_package with the vcpkg toolchain:

find_package(sablib CONFIG REQUIRED)
target_link_libraries(my_app PRIVATE sablib::sablib)

Via Conan

After installing with Conan, use find_package with the Conan toolchain:

find_package(sablib CONFIG REQUIRED)
target_link_libraries(my_app PRIVATE sablib::sablib)

Manually

You can add sablib to your project as follows:

# In your CMakeLists.txt
set(SABLIB_DIR "/path/to/sablib")
include_directories(${SABLIB_DIR})
find_library(SABLIB_LIB NAMES sablib libsablib PATHS "${SABLIB_DIR}/build")
target_link_libraries(my_app PRIVATE ${SABLIB_LIB} Eigen3::Eigen)

Ensure Eigen3 is configured in your project (e.g., via find_package(Eigen3 REQUIRED)).

Header Inclusion

Include the main header to access all library features:

#include "sablib/sablib.h"

Alternatively, you can include specific algorithm headers to minimize dependencies:

#include "sablib/smoothing/pspline.h"
#include "sablib/baseline/arpls.h"

Documentation

Please see the GitHub Pages for documentation.

License

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