Good morning everyone. I have a large dataset of photographs similar to the one attached. It would be my intention to train a neural network to recognise water and "non-water" zones. How can I proceed? Thank you
$\begingroup$
$\endgroup$
2
-
$\begingroup$ Have you tried something? If not, you need to add details on what you know, what you don't, so that we can provide relevant pointers. A starting point could be the documentation (such as this example: reference.wolfram.com/language/tutorial/…). Or are you looking for someone to do the job for you? $\endgroup$– anderstoodCommented Feb 23, 2023 at 17:36
-
$\begingroup$ My colleagues developed a programme in Python based on a supervised CNN in two steps. 1) Segmentation of the image to obtain a binary mask of water and not water 2) Analysis of the mask to deduce the position of the waterline. The dynamic nature of water can produce reflections that often fool the system. It is mounted on a multifunctional robotic boat. My purpose with this post was to stimulate a discussion and to see if, with Mathematica's powerful libraries, something could be done better and faster. Thank you to those who contribute constructively to this post. $\endgroup$– Ramiro dell'ErbaCommented Feb 28, 2023 at 8:30
Add a comment
|