I am currently doing my master thesis on “Dark feature segmentation for oil spill detection using SAR imagery”.
Currently I am in the middle of my literature review for the most robust segmentation algorithms and I would love if anyone could point me in the right direction. Should I go to the more classical adaptive threshold algorithms or go to a more modern approach using neural networks.
In addition, I would like to know what is the exact algorithm that is used in SNAP for oil-spill segmentation. I know is based on adaptive threshold but I would like to study in detail the actual algorithm.
Thanks in advance!
Ps: links to the actual research papers/tools/libraries would be very much appreciated
Isn’t this kind of strategical question something that you better discuss with your supervisor? Surely, people could have an opinion on this, but I’m afraid an open discussion might lead you somewhere away from the topic you both agreed upon. Please correct me if I misunderstood your question.
On the implementation of the Oil Spill in SNAP, one of the develpers will be able to help you, of course.
That could be truth in the case that my supervisor was an expert in the remote sensing field, but he is not. That is why I wanted to get the opinion of professionals that have experience in the area. I do not see how asking for a general advice to experienced people in remote sensing could be misleading for the research section of my master thesis. I need to assess all options so the more information I get the better. Plus I am still fairly new to the field myself.