I do not find the actual legend for the color-coding in the output colored images. According to the example in HooverCompareSegmentation — Orfeo ToolBox 8.1.2 documentation,
I guess:
green: correct detection
red: missed
purple: over-segmented
cyan: under-segmented
but I would like a confirmation.
Also, this output does not seem to match the one described in
https://www.orfeo-toolbox.org/CookBook/C++/Examples/OBIA/HooverMetricsEstimation.html#hoovermetricsestimation-cxx
// The output image contains for each GT region its correct detection score (“RC”, band 1),
// its over-segmentation score (“RF”, band 2), its under-segmentation score (“RA”, band 3)
// and its missed detection score (“RM”, band 4).std::cout << “Mean RC =” << instances->GetMeanRC() << std::endl;
std::cout << “Mean RF =” << instances->GetMeanRF() << std::endl;
std::cout << “Mean RA =” << instances->GetMeanRA() << std::endl;
std::cout << “Mean RM =” << instances->GetMeanRM() << std::endl;
std::cout << “Mean RN =” << instances->GetMeanRN() << std::endl;// The Hoover scores are also computed for the whole segmentations. Here is some explanation about the score names :
// C = correct, F = fragmentation, A = aggregation, M = missed, N = noise.