double_c_data

double_c_data, a Python code which creates, plots, or saves a double C dataset.

The data is confined to two separate regions, each having the shape of the letter "C". The two regions are close, and interlocking, but do not actually touch. This data represents a difficult challenge for algorithms that attempt to detect natural segmentations of data.

Licensing:

The information on this web page is distributed under the MIT license.

Languages:

double_c_data is available in a MATLAB version and an Octave version and a Python version.

Related Data and Programs:

components, a Python code which seeks the connected "nonzero" or "nonblack" components of an image or integer vector, array or 3D block.

random_data, a Python code which uses a random number generator (RNG) to sample points corresponding to various probability density functions (PDF), spatial dimensions, and geometries, including the annulus, circle, ellipse, ellipsoid, hypercube, hypersphere, simplex, tetrahedron and triangle.

sammon_data, a Python code which generates six sets of M-dimensional data for cluster analysis.

Source Code:

• double_c_data.py, the source code.
• double_c_data.sh, runs all the tests.
• double_c_data.txt, the output file.

• double_c_data_800_600.png, a plot of the data for N0 = 800, N1 = 600.
• double_c_data_100_100.txt, a data file with N0 = 100, N1 = 100.
• double_c_data_100_400.txt, a data file with N0 = 100, N1 = 400.
• double_c_data_25_25.txt, a data file with N0 = 25, N1 = 25.
• double_c_data_500_500.txt, a data file with N0 = 500, N1 = 500.