Connected component algorithms are basic building blocks of many computer vision algorithms. With bwconncomp, MatLab provides a simple connected components algorithm for binary images. Motivated by the need to compute connected components in (semantic) segmentations, where applying bwconncomp is painful, I wrote a MEX wrapper for Ali Rahimi's C++ implementation of a connected component algorithm.

The code is available on GitHub:

Building

The MEX wrapper is compiled as follows:

>> mex sp_fast_connected_relabel.cpp
Building with 'g++'.
Warning: You are using gcc version '4.8.4'. The version of gcc is not supported. The version currently supported with MEX is
'4.7.x'. For a list of currently supported compilers see: http://www.mathworks.com/support/compilers/current_release. 

Warning: You are using gcc version '4.8.4-2ubuntu1~14.04.1)'. The version of gcc is not supported. The version currently
supported with MEX is '4.7.x'. For a list of currently supported compilers see:
http://www.mathworks.com/support/compilers/current_release. 

MEX completed successfully.

Usage

The wrapper operates on matrices of type double. A usage example is given below:

>> image = imread('checkerboard.png');
>> labels = sp_fast_connected_relabel(double(image));
>> imshow(uint8(labels)*10)
>> imwrite(uint8(labels)*10, 'checkerboard_components.png');

Given an image checkerboard.png, Figure 1 shows the generated output.