Lime (Library for IMage Editing) is an OpenCV-dependent image editing library for C++. Since this library is comprised of only headers, you can use it simply by including 'lime.hpp'.

This library is comprised of only header files. Therefore, you do not need to do anything other than installing OpenCV in your computer. The library depends on OpenCV 3.0 or later.

For the installation detail, please refer to the reference manual.

This library supports following methods.

• Horn's Algorithm
• Rahman's Algorithm
• Faugeras's Algorithm

• NPR edge detector with Difference of Gaussian (XDoG and FDoG supported)
• NPR image filters
  • Anisotropic diffusion
  • Mean curvature flow
  • Shock filter
  • Kuwahara filter (and its variations)
  • Mathematical morphology
• Line integral convolution
• Vector field detection
  • Smoothed stractural tensor
  • Edge tangent field
  • Vector field singularity detector
• Poisson disk sampling
  • Simple random-queue-based sampler
  • Fast parallel sampler
• Noise generation
  • Uniform noise
  • Perlin noise

• Mean squared error (MSE)
• Root mean squared error (RMSE)
• Peak signal-to-noise ratio (PSNR)
• Structural similarity (SSIM)
• Multi-scale structural similarity (MS-SSIM)
• Color multi-scale structural similarity (CM-SSIM)

You can compile all the samples using CMake!

MIT License 2015-2019, Tatsuya Yatagawa (tatsy)