Open melisandeteng opened 4 years ago
The method of dealing with such problems, in general, would be to use a combination of the distance transform of the binary marks and followed by morphological erosion and dilation. The best function implementation to do these tasks would be
There are two main things for which postprocessing steps could be used in ground/flood masks obtained with Deeplab models:
Ideas for criteria
If it is too big, the postprocessing step might lead to even more erroneous masks. Example below: ground segmentation with Deeplabv3 and ground segmentation with the postprocessing step.