This repository contains python and cython codes that can generate noise images, which can be used for texture and heightmap to visualize the terrain in 3D. In the python modules, numpy, and in the Cython modules, C array is mainly used. Those modules have the same functions, which return the array to be converted to an image. Their difference is speed. See speed comparison result below.
Using the texture_generator's methods, textures such as clouds and cubemaps can be procedurally created. See texture_generator.
- Cython 3.0.12
- numpy 2.2.4
- opencv-contrib-python 4.11.0.86
- opencv-python 4.11.0.86
- Python 3.12
- Windows11
git clone --recursive https://github.com/taKana671/NoiseTexture.git
cd NoiseTexture
python setup.py build_ext --inplace
If the error like "ModuleNotFoundError: No module named ‘distutils’" occurs, install the setuptools.
pip install setuptools
from cynoise.perlin import PerlinNoise
# from pynoise.perlin import PerlinNoise
from output_image import output_image_8bit, output_image_16bit
maker = PerlinNoise()
arr = maker.noise3()
output_image_8bit(arr)
output_image_16bit(arr)
# change the number of lattices and the image size. The grid default is 4, size default is 256.
maker = PerlinNoise(size=512, grid=8)
A noise image is output as png file.
For more details of methods and parameters, please see source codes.
The execution time of each methods were measured like below.
maker = VoroniNoise()
result = %timeit -o maker.noise2()
print(result.best, result.loops, result.repeat)
| python | cython | |||||
|---|---|---|---|---|---|---|
| method | best(s) | loops | repeat | best(s) | loops | repeat |
| PerlinNoise.noise2 | 1.258439 | 1 | 7 | 0.016928 | 10 | 7 |
| PerlinNoise.noise3 | 2.113114 | 1 | 7 | 0.023845 | 10 | 7 |
| PerlinNoise.fractal2 | 5.185004 | 1 | 7 | 0.047277 | 10 | 7 |
| PerlinNoise.warp2_rot | 21.5.049 | 1 | 7 | 0.167894 | 1 | 7 |
| PerlinNoise.wrap2 | 20.60505 | 1 | 7 | 0.162479 | 10 | 7 |
| CellularNoise.noise2 | 1.772891 | 1 | 7 | 0.034365 | 10 | 7 |
| CellularNoise.noise3 | 4.445742 | 1 | 7 | 0.076830 | 10 | 7 |
| CellularNoise.noise24 | 5.562702 | 1 | 7 | 0.089216 | 10 | 7 |
| CellularNoise.cnoise2 | 5.574327 | 1 | 7 | 0.146625 | 10 | 7 |
| CellularNoise.cnoise3 | 15.21613 | 1 | 7 | 0.330184 | 1 | 7 |
| PeriodicNoise.noise2 | 1.511534 | 1 | 7 | 0.017240 | 100 | 7 |
| PeriodicNoise.noise3 | 2.522443 | 1 | 7 | 0.023741 | 10 | 7 |
| VoronoiNoise.noise2() | 1.782102 | 1 | 7 | 0.113907 | 10 | 7 |
| VoronoiNoise.noise3() | 4.200216 | 1 | 7 | 0.179295 | 10 | 7 |
| VoronoiEdges.edge2() | 16.96586 | 1 | 7 | 0.195019 | 1 | 7 |
| VoronoiEdges.edge3() | 83.08460 | 1 | 7 | 1.026584 | 1 | 7 |
| VoronoiRoundEdges.round2() | 18.23457 | 1 | 7 | 0.245364 | 1 | 7 |
| VoronoiRoundEdges.round3() | 84.45329 | 1 | 7 | 1.233588 | 1 | 7 |
| SimplexNoise.noise2 | 1.656967 | 1 | 7 | 0.020974 | 10 | 7 |
| SimplexNoise.noise3 | 4.337698 | 1 | 7 | 0.024148 | 10 | 7 |
| SimplexNoise.fractal2 | 6.711880 | 1 | 7 | 0.065214 | 10 | 7 |
| SimplexNoise.fractal3 | 17.35970 | 1 | 7 | 0.083178 | 10 | 7 |
| ValueNoise.noise2 | 1.128080 | 1 | 7 | 0.016823 | 100 | 7 |
| ValueNoise.noise3 | 1.566021 | 1 | 7 | 0.022034 | 10 | 7 |
| ValueNoise.grad2 | 3.875698 | 1 | 7 | 0.034712 | 10 | 7 |
| ValueNoise.fractal2 | 3.839937 | 1 | 7 | 0.041812 | 10 | 7 |
| ValueNoise.warp2_rot | 16.90730 | 1 | 7 | 0.151481 | 10 | 7 |
| ValueNoise.warp2 | 15.49489 | 1 | 7 | 0.142454 | 10 | 7 |