@@ -39,9 +39,9 @@ Mke.activate!(type = "png")
3939
4040We will use an image from the
4141[ Harmonized Sentinel-2 MSI] ( https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_S2_SR_HARMONIZED )
42- dataset. The image is the result of cloud masking and averaging between
43- dates 2025-07-01 and 2025-07-30, followed by a download within a specified
44- box of latitude and longitude values :
42+ dataset, downloaded within a box of latitude and
43+ longitude values after cloud removal and averaging
44+ between dates 2025-07-01 and 2025-07-30 :
4545
4646``` {julia}
4747image = GeoIO.load("data/sentinel.tif")
@@ -96,10 +96,6 @@ tasks. Second, we would like to convert the region of the image that is inside
9696of the agricultural field into a polygonal area for measurements of area and
9797perimeter.
9898
99- The ability to convert geospatial data between these two representations (i.e.,
100- "raster" vs. "vector") based on a categorical variable is key for advanced
101- geospatial data science workflows.
102-
10399## Methodology
104100
105101In order to assign meaning to the pixels of the image, we first need to visualize
@@ -142,7 +138,7 @@ We can map the function to all the pixels of the image using the `Map` transform
142138color = rgb |> Map(["R", "G", "B"] => ascolor => "RGB")
143139```
144140
145- The ` viewer ` displays the colors that are familiar to us :
141+ The ` viewer ` displays the colors and reveals the features of the land :
146142
147143``` {julia}
148144color |> viewer
219215### Image segmentation
220216
221217From visual inspection, we select the ` "MGRVI" ` index for image segmentation.
222- It creates a strong contrast between pixels that are inside and outside the
218+ It produces a strong contrast between pixels that are inside and outside the
223219agricultural field. We use the ` Map ` transform again to create a binary
224220variable in terms of the 30th percentile of the spectral index:
225221
@@ -234,7 +230,7 @@ binary |> viewer
234230```
235231
236232Additionally, we use the ` ModeFilter ` transform to eliminate small artifacts
237- that are not relevant for the stated objectives:
233+ that are not relevant for our stated objectives:
238234
239235``` {julia}
240236mask = binary |> ModeFilter()
0 commit comments