2
GeoServer provides rendering transformations that can do exactly this. The manual contains an example using Jiffle that calculates NDVI on the fly from multiband Sentinel data.
<Transformation>
<ogc:Function name="ras:Jiffle">
<ogc:Function name="parameter">
<ogc:Literal>coverage</ogc:Literal>
...
2
Here is a code i adapted for you, it worked on my qgis 2.18.28:
b = iface.activeLayer()
fni = b.fieldNameIndex("rgb")
unique_values = b.uniqueValues(fni)
# fill categories
categories = []
for unique_value in unique_values:
red, blue, green = unique_value.split(",")
# initialize the default symbol for this geometry type
...
2
You could try to define visualization parameters and then make a new variable where you add the visualization. It would look something like this:
var vis = {
min: 0,
max: 3000,
bands:['B4','B3','B2'],
};
var ffa_db_new = ffa_db.visualize(vis);
And then you export the new variable ffa_db_new instead of the old one.
2
Having only RGB you cannot calculate the NDVI. As stated in the comment, you need as well the NIR (near-infrared) channel (https://gisgeography.com/ndvi-normalized-difference-vegetation-index/)
1
Try to build a Virtual Raster. Make sure to check Place each input file into a separate band
1
Looks like you haven't defined a variable called image - which is what you pass along to the visualize() function. Try changing your visualize line to:
var imageRGB = dataset.median().visualize({bands: ['B4', 'B3', 'B2'], min: 0, max: 2500});
print("rgb",imageRGB);
You’ll also need to update the geometry in your export because ‘image’ doesn’t ...
1
The palr package has a function image_raster for the task of baking in the palette to RGB version of the raster. (You can set the extent and projection using raster itself as usual).
library(raster)
#> Loading required package: sp
r <-raster(volcano)
library(palr)
## apply the colours used by raster plot() itself
rgb0 <- image_raster(r, col = ...
Only top voted, non community-wiki answers of a minimum length are eligible
