Need to identify human settlement areas from Landsat satellite images? Follow these steps using ArcGIS Pro.

“Human settlements maps are useful in understanding growth patterns, population distribution, resource management, change detection, and a variety of other applications where information related to earth surface is required. Human settlements classification is a complex exercise and is hard to capture using traditional means. Deep learning models are highly capable of learning these complex semantics and can produce superior results.” Source.


(1) Necessary Downloads: Sample Output

(2) Prepare Data in a Working Directory

  • Create or identify an existing file geodatabase in a working directory
  • Copy the two tools (Manage Landsat 8 imagery & Human Settlements Classification (Landsat 8)) to the working directory
  • Copy the Landsat Collection 1 Level 1 data to the working directory

(3) Create a multispectral mosaic

  • In the Catalog Pane, browse to the landsat8.pyt tool and launch landsat8
    • Type in the Mosaic Dataset Name if creating a new mosaic dataset or pick an existing one
    • Select the input Folder where the Landsat data is stored.
    • Select the Landsat level for the input data. Based on the level of the data additional checkbox is updated
    • Click on any additional check boxes as desired to created additional products via referenced mosaic datasets.
    • Click OK to continue.

(4) Classify Human Settlements

(5) Convert to feature class

  • Use the Raster to Polygon tool to convert
Deep Learning to Identify Human Settlements from Landsat Imagery
Tagged on:

Leave a Reply

Your email address will not be published. Required fields are marked *