Scikit Image Clustering Scripts
Python scripts using scikit-image and scikit-learn to cluster images.
files
- basics/otsu_thresholding.py: Performs the Otsu thresholding and show the original gray-scale image and the binarized image.
- basics/histogram.py: Calculates & show the histogram of the gray scale image.
- basics/morphology.py: Dilates/Erodes/Closes/Opens the binary thresholded image.
- basics/equalization.py: Thresholding a histogram-equalized image (enhanced in contrast).
- basics/difference.py: performs arithmetic difference between two images.
- preprocessing/adaptive_filters.py: Compare adaptive filters (lee, frost, kuan) to mean/median filters.
- clustering/kmeans_clustering.py: Performs a k-means clustering followed by an Otsu thresholding.
- clustering/kmeans_spatial.py: Performs a k-means clustering taking into account the spatial context (X, Y).
- clustering/fuzzy_c_means.py: Clustering using Fuzzy C-means algorithm.
- clustering/gaussian_mixture_model.py: Gaussian Mixture Model on image population.
- clustering/lbp_texture.py: Local Binary Pattern texture segmentation.
- clustering/mean_shift.py: Performs the Mean Shift algorithm on the image.
- segmentation/flood_fill.py: 8-neighbours flood fill with prior initialization.
- segmentation/fill_holes.py: Fill both black and white holes.
- segmentation/local_maxima.py: Detect objects using local maxima.
Prerequisites
All the followding packages can be installed on Windows with conda install <package>:
- numpy & scipy:
pip install numpy scipy
- scikit-image:
pip install scikit-image
- scikit-learn:
pip install scikit-learn
- scikit-fuzzy:
pip install scikit-fuzzy
- pyradar:
pip install pyradar
- gdal-gdal:
apt-get install python-gdal
