The simple region growing method proved to be good
enough for the initial segmentation of pigmented skin lesions
images. For all of the images used the implemented method
offered better results than the non-uniform HSV quantization
method. The low computational complexity of the method
makes it suitable for smartphone or web aplications. In terms
of color space the results proved that using the RGB color
space for image representation and segmentation leads to good
segmented regions so that no color space transformation will
be necessary in a risk evaluation aplication. Regarding the
quality and resolution of the images the results revealed that if
the images are taken in reasonable daylight lighting conditions
there is no preferred resolution value. By extracting the region
of interest from the cutaneous pigmented image we
significantly decrease the computational complexity required
further for the following image processing operations used for
implementation of the ABCDE rule.