This Leaflet plugin provides multispectral channel manipulation and processing tools (such as NDVI or other remote sensing methods) for Leaflet tile layers using pure client-side JavaScript. It uses image-sequencer and was developed by Public Lab.
It is based on the simpler plugin for ImageOverlays at https://github.com/publiclab/leaflet-multispectral
It is available as a node module as leaflet-tile-filter
.
This library was made possible in part by NASA's AREN project.
See the demo for basic usage: https://publiclab.github.io/leaflet-tile-filter/
// create an tile layer
var tileLayer = L.tileLayer(
'https://gibs-{s}.earthdata.nasa.gov/wmts/epsg3857/best/' +
'{layer}/default/{time}/{tileMatrixSet}/{z}/{y}/{x}.jpg',
{
layer: 'MODIS_Terra_CorrectedReflectance_Bands721',
tileMatrixSet: 'GoogleMapsCompatible_Level9',
maxZoom: 9,
time: '2013-11-04',
tileSize: 256,
subdomains: 'abc',
noWrap: true,
continuousWorld: true,
// Prevent Leaflet from retrieving non-existent tiles on the
// borders.
bounds: [
[-85.0511287776, -179.999999975],
[85.0511287776, 179.999999975]
],
attribution:
'<a href="https://wiki.earthdata.nasa.gov/display/GIBS">' +
'NASA EOSDIS GIBS</a> ' +
'<a href="https://github.com/nasa-gibs/web-examples/blob/master/examples/leaflet/webmercator-epsg3857.js">' +
'View Source' +
'</a>'
});
// apply a colormap:
tileLayer.filter('colormap');
tileLayer.addTo(map);
To revert the image to it's previous state, use:
tileLayer.revert();
With a small change coming in Image Sequencer, multi-step commands and the full Image Sequencer syntax will be available:
// apply NDVI equation to the image, and then a colormap:
tileLayer.filter('ndvi,colormap');
For more complex commands, you can use JavaScript expresssions, in the following format:
tileLayer.filter('dynamic{red:R*2|green:B|blue:B/2}');
In this example, we're using Image Sequencer's dynamic
module to set the displayed RGB values individually, with the expressions R*2
, B
, and B/2
, respectively.
(coming soon) Commands you can pass into the filter are extremely flexible; they are Image Sequencer string syntax (full object notation coming soon). Read about Image Sequencer and how to use it here:
https://github.com/publiclab/image-sequencer
Using the visual editor, you can develop a command string to pass into this filter:
https://sequencer.publiclab.org
Image Sequencer is implemented in pure JavaScript, and is under development; we expect optimizations as well as worker threads and WebAssembly to improve performance in upcoming versions.
Learn about multi-band imagery, some of it's uses at this great blog post by Charlie Lloyd of Mapbox:
https://blog.mapbox.com/putting-landsat-8s-bands-to-work-631c4029e9d1
Here's some sources of Landsat 8 data:
Here's more on NDVI and DIY techniques to collect multi-band imagery: https://publiclab.org/infragram