Classify your city (summary)

[Paul Alexander]

Deriving a LCZ classification using the WUDAPT method is very straightforward, but requires an understanding of the LCZ classification scheme and local knowledge of your city of interest.

We have divided it into 3 simple steps, for creating training areas, you should watch this video

•Step 1: Download the data and software
•Step 2: Generate the LCZ classification in SAGA GIS
•Step 3: Upload your files

You can upload your KMZ training areas from google earth at this location

• This topic was modified 8 years, 2 months ago by Paul Alexander.
• This topic was modified 8 years, 2 months ago by Paul Alexander.

[Donny]

hi,

I already join WUDAPT, and download SAGA software,
when I follow step by step how to make classificarion using LCZC
but I did not find LCZC tool in my SAGA software, do I need install another plugin?

Thnak you

[Paul Alexander]

Donny,

Thanks for your interest in WUDAPT apologies for the delay in replying to you!

You should see the Local Climate Zone Classification tool under Geoprocessing -> Imagery-> Classification

Did you download Saga 2.2.0 from here?

We will be happy to help you out once we get some more information from you 🙂

Paul.

[Marie-Leen Verdonck]

Hi,

I have the same problem. I can not find the Local Climate Zone Classification tool and I downloaded Saga 2.2.0 from the link you provided.

Best regards,
Marie-Leen

[Benni]

That is strange. Try the following. Save this file to your local computer. Open it in SAGA with -> geoprocessing -> Load tool library.

Can you see the tool now?

[marieleen verdonck]

Yes!
Thank you

[Marie-leen Verdonck]

Hi,

When running the tool “OGR: import vector data” there seems to be a problem. The tool won’t execute, I solved this problem for importing the ROI. But the LCZ classifcation tool won’t run when the training areas (also kmz file) are not read properly. Does anybody else have this problem as well?

Marie-Leen

[Alvin]

Is there a limitation to the ROI size/area.
I have created a training dataset but when I run the LCZ classification for a region more than half of the landsat area I get the error, “insufficient training samples”.
Are there other reasons aside from the insufficient training sample?

[Author]

Posts