How do I extract pixels of specific values from a raster in Google Earth Engine?

Question

Say in this example, the classified results are: water (2), vegetation (1) and urban (0). I want to extract water only, i.e. pixels with the value 2. I couldn't find a function in GEE that would let me specifically extract out pixels of a certain value, so I tried to carry out the following:

• utilised .lte() function to extract pixels of value 1 and 0 i.e. not water
• mask out non-water pixels (i.e. 1 and 0) from the classified results such that only water pixels will show

However, it failed and I'm not sure where I went wrong. So my questions are as follows:

1. Where did I go wrong with the above workflow?
2. Is there a more straightforward way to directly extract pixels of a specific value, say in this case 2, without having to go through the hassle of masking etc? What if I wanted pixels within a certain range, say 1-2? Would it be the same function?

Many thanks.

// Make a cloud-free Landsat 8 TOA composite (from raw imagery).
var l8 = ee.ImageCollection('LANDSAT/LC08/C01/T1');

var image = ee.Algorithms.Landsat.simpleComposite({
  collection: l8.filterDate('2018-01-01', '2018-12-31'),
  asFloat: true
});

// Use these bands for prediction.
var bands = ['B2', 'B3', 'B4', 'B5', 'B6', 'B7', 'B10', 'B11'];

// Load training points. The numeric property 'class' stores known labels.
var points = ee.FeatureCollection('GOOGLE/EE/DEMOS/demo_landcover_labels');

// This property stores the land cover labels as consecutive
// integers starting from zero.
var label = 'landcover';

// Overlay the points on the imagery to get training.
var training = image.select(bands).sampleRegions({
  collection: points,
  properties: [label],
  scale: 30
});

// Train a CART classifier with default parameters.
var trained = ee.Classifier.smileCart().train(training, label, bands);

// Classify the image with the same bands used for training.
var classified = image.select(bands).classify(trained);


var random = classified.lte(1);
//extracting out pixels less than or equals to 1, i.e. not water
var notWater = classified.updateMask(random);

var onlyWater = classified.updateMask(notWater);

// Display the inputs and the results.
Map.centerObject(points, 11);
Map.addLayer(image, {bands: ['B4', 'B3', 'B2'], max: 0.4}, 'image');
Map.addLayer(notWater,
             {},
             'not water');
Map.addLayer(classified,
             {min: 0, max: 2, palette: ['red', 'green', 'blue']},
             'classification');
Map.addLayer(random,
             {},
             'random');
Map.addLayer(classified.updateMask(random),
             {},'classified masked with random');

// to display only water pixels 
Map.addLayer(onlyWater,
             {},
             'only water');

Answer 1

I'm not completely sure this is exactly what you're asking for:

var onlyWater = classified.updateMask(
  classified.eq(2) // Only keep pixels where class equals 2
)

var notWater = classified.updateMask(
  classified.neq(2) // Only keep pixels where class not equals 2
)

If you need to define a range instead, you could do this (doesn't really make sense for this use case, if I get you right, but I'll show it anyway):

var onlyWater = classified.updateMask(
  classified.gt(1).and(classified.lt(3))
)

// eq() - equals
// neq() - not equals
// lt() - less than 
// gt() - greater than
// lte() - less than or equals
// gte() - greater than or equals
// and() - logical and
// or() - logical or