georaster

Wrapper around Georeferenced Rasters like GeoTIFF, NetCDF, JPG, and PNG that provides a standard interface. You can also create your own georaster from simple JavaScript objects.

load from url on front-end

const parseGeoraster = require("georaster");
fetch(url)
  .then(response => response.arrayBuffer() )
  .then(parseGeoraster)
  .then(georaster => {
      console.log("georaster:", georaster);
  });

load from file on back-end

const parseGeoraster = require("georaster");
fs.readFile("data/GeogToWGS84GeoKey5.tif", (error, data) => {
    parseGeoraster(data).then(georaster => {
      console.log("georaster:", georaster);
    })
});

load from simple object on front-end

const parseGeoraster = require("georaster");
const values = [ [ [0, 1, 2], [0, 0, 0], [2, 1, 1] ] ];
const noDataValue = 3;
const projection = 4326;
const xmin = -40;
const ymax = 14;
const pixelWidth = 0.00001;
const pixelHeight = 0.00001;
const metadata = { noDataValue, projection, xmin, ymax, pixelWidth, pixelHeight };
const georaster = parseGeoraster(values, metadata);

load cloud optimized geotiff

This option allows you to basically load the pixels only when you need them versus the other options that require you to load the whole image into memory. It will also attempt to automatically discover any available overview files.

where to clip

top is how many pixels from the top of the image to skip over before we start clipping

bottom is how many pixels from the bottom of the image to skip over before we start clipping

left is how many pixels in from the left side of the image to skip over before we start clipping

right is how many pixels in from the right side of the image to skip over before we start clipping

clipping resolution

width is how many pixels wide should be the returned image. This helps to configure the resolution.

height is how many pixels tall should be the returned image. This helps to configure the resolution.

resampling

resampleMethod is how to resample the pixels for the returned image. This value is passed on to geotiff.js' readRasters function and defaults to 'bilinear'. The alternative is 'nearest', which is faster and better for categorical data like landcover and continous classes with smooth variation like temperature.

  const raster_url = "https://landsat-pds.s3.amazonaws.com/c1/L8/024/030/LC08_L1TP_024030_20180723_20180731_01_T1/LC08_L1TP_024030_20180723_20180731_01_T1_B1.TIF";
  parseGeoraster(raster_url).then(georaster => {
    console.log(georaster.height);
    const options = { left: 0, top: 0, right: 4000, bottom: 4000, width: 10, height: 10 };
    georaster.getValues(options).then(values => {
      console.log("clipped values are", values);
    });
  });

required properties

name	description
maxs	array with max value for each band
mins	array with min value for each band
ranges	array with difference between max and min value for each band
noDataValue	no data value
pixelWidth	width of pixel in dimension of coordinate reference system
pixelHeight	height of pixel in dimension of coordinate reference system
projection	equal to EPSG code, like 4326
values	two dimensional array of pixel values (for Cloud Optimized GeoTIFF's, use getValues instead)
width	number of pixels wide raster is
xmax	xmax in crs, which is often in longitude
xmin	xmin in crs, which is often in longitude
ymin	ymin in crs, which is often in latitude
ymax	ymax in crs, which is often in latitude

optional properties

name	description
palette	Array that maps raster values to RGBA colors

functions

name	description
getValues	described above
toCanvas	experimental! returns a canvas picture of the data. You can pass in options object with height or width specified

loading georaster package through a script tag

<script src="https://unpkg.com/georaster"></script>

You can view a simple demo of this here

Support

Post a Github issue or contact the package author, Daniel J. Dufour, at daniel.j.dufour@gmail.com