[x] Convert the 3 single masks to a single multi-class mask where 1 is BuiltUp, 2 is NonBuiltUp and 3is Slum. Visualize the ground truth indexed image (pure and overlaid on the original image).
[x] Visualize overlaid ground truth and segmentation on top of original image
[x] Fill missing pixels segmentation. Visualize the resulting segmentation image (pure and overlaid on the original image).
[x] De-noise (via majority filter?) the segmentation. Visualize the resulting segmentation image (pure and overlaid on the original image).
[ ] Make a script to compare ground truth to partial segmentation
[ ] Compare interpolated segmentation to ground truth
For implementing majority filter perhaps this will work:
fun = @(block_struct) mode(block_struct.data(:)) * ones(size(block_struct.data)); output_matrix = blockproc(input_matrix,[<size_x> <size_y>],fun);
[x] Convert the 3 single masks to a single multi-class mask where 1 is BuiltUp, 2 is NonBuiltUp and 3is Slum. Visualize the ground truth indexed image (pure and overlaid on the original image).
[x] Visualize overlaid ground truth and segmentation on top of original image
[x] Fill missing pixels segmentation. Visualize the resulting segmentation image (pure and overlaid on the original image).
[x] De-noise (via majority filter?) the segmentation. Visualize the resulting segmentation image (pure and overlaid on the original image).
[ ] Make a script to compare ground truth to partial segmentation
[ ] Compare interpolated segmentation to ground truth