Detect masonry from an image

Hi

I'm trying to create a file which detects masonry from walls (eg. facade of a building). So basically this code should detect masonry from an image and isolate it into a 2D model. What's the best way to do this?

Thanks in advance.

-Ax

Answer 1

[Image Analyst]

Depends on how varied your images are.

1. If it's any possible kind of building (wood, glass, steel, masonry, etc.) and any possible kind of masonry (brick, blocks, marble, limestone, stucco, cement, tile, different colors, etc.) then I think it could be difficult. You'd probably have to use SegNet to train a deep learning model with thousands of images.
2. If you have a more limited set of images, like houses made of only red brick, you could possibly get by with traditional methods like a texture filter like stdfilt() or entropyfilt().

[Axel Reyskens]

I think option one applies. It could be any building and any type of masonry. So I'm trying to 'detect' masonry from an image of a facade of a building. So the masonry could be any type and even colours can differ (eg. red masonry vs dark masonry).

[Axel Reyskens]

Is using image segmentation an option?

Also: if I want to make a 2D model out of it later on, how can I identify length?