Dynamic thresholding ftw
Find the scientific description of the algorithm in the paper of Markert et al. (2020). Comparing Sentinel-1 Surface Water Mapping Algorithms and Radiometric Terrain Correction Processing in Southeast Asia Utilizing Google Earth Engine
Figure 2 shows the workflow for the data processing applied.
Copy the code below or use this link
//Script for SAR water segmentation with Edge Otsu Algorithm in Cambodia
//Slope Correction Algorithms
// Implementation by Andreas Vollrath (ESA), inspired by Johannes Reiche (Wageningen)
function slopeCorrection(image) {
var imgGeom = image.geometry()
var srtm = ee.Image('JAXA/ALOS/AW3D30/V2_2').select('AVE_DSM').clip(imgGeom)
var sigma0Pow = ee.Image.constant(10).pow(image.divide(10.0))
// Article ( numbers relate to chapters)
// 2.1.1 Radar geometry
var theta_i = image.select('angle')
var phi_i = ee.Terrain.aspect(theta_i)
.reduceRegion(ee.Reducer.mean(), theta_i.get('system:footprint'), 1000)
.get('aspect')
// 2.1.2 Terrain geometry
var alpha_s = ee.Terrain.slope(srtm).select('slope')
var phi_s = ee.Terrain.aspect(srtm).select('aspect')
// 2.1.3 Model geometry
// reduce to 3 angle
var phi_r = ee.Image.constant(phi_i).subtract(phi_s)
// convert all to radians
var phi_rRad = phi_r.multiply(Math.PI / 180);
var alpha_sRad = alpha_s.multiply(Math.PI / 180);
var theta_iRad = theta_i.multiply(Math.PI / 180);
var ninetyRad = ee.Image.constant(90).multiply(Math.PI / 180);
// slope steepness in range (eq. 2)
var alpha_r = (alpha_sRad.tan().multiply(phi_rRad.cos())).atan();
// slope steepness in azimuth (eq 3)
var alpha_az = (alpha_sRad.tan().multiply(phi_rRad.sin())).atan();
// local incidence angle (eq. 4)
var theta_lia = (alpha_az.cos().multiply((theta_iRad.subtract(alpha_r)).cos())).acos();
var theta_liaDeg = theta_lia.multiply(180 / Math.PI);
// 2.2
// Gamma_nought_flat
var gamma0 = sigma0Pow.divide(theta_iRad.cos())
var gamma0dB = ee.Image.constant(10).multiply(gamma0.log10());
var ratio_1 = gamma0dB.select('VV').subtract(gamma0dB.select('VH'));
// Volumetric Model
var nominator = (ninetyRad.subtract(theta_iRad).add(alpha_r)).tan();
var denominator = (ninetyRad.subtract(theta_iRad)).tan();
var volModel = (nominator.divide(denominator)).abs();
// apply model
var gamma0_Volume = gamma0.divide(volModel);
var gamma0_VolumeDB = ee.Image.constant(10).multiply(gamma0_Volume.log10());
// we add a layover/shadow maskto the original implmentation
// layover, where slope > radar viewing angle
var alpha_rDeg = alpha_r.multiply(180 / Math.PI);
var layover = alpha_rDeg.lt(theta_i);
// shadow where LIA > 90
var shadow = theta_liaDeg.lt(85);
// calculate the ratio for RGB vis
var ratio = gamma0_VolumeDB.select('VV').subtract(gamma0_VolumeDB.select('VH'));
var output = gamma0_VolumeDB.addBands(ratio).addBands(alpha_r).addBands(phi_s).addBands(theta_iRad)
.addBands(layover).addBands(shadow).addBands(gamma0dB).addBands(ratio_1);
return image.addBands(
output.select(['VV', 'VH', 'slope_1', 'slope_2'], ['VV', 'VH', 'layover', 'shadow']),
null,
true
).addBands(image.select("angle"));
}
/***********End of Edge Otsu************/
/**
* Perona-Malik (anisotropic diffusion) convolution
*
* by Gennadii Donchyts see https://groups.google.com/forum/#!topic/google-earth-engine-developers/a9W0Nlrhoq0
* I(n+1, i, j) = I(n, i, j) + lambda * (cN * dN(I) + cS * dS(I) + cE * dE(I), cW * dW(I))
*
* I: ee.Image single band, natural units
* iter: Number of interations to apply filter
* K: kernal size
* opt_method: choose method 1 (default) or 2, DETAILS
*
* Returns: single band ee.Image in natural units
*
* Example: image = PeronaMalik(image, 10, 3.5, 1)
*/
function PeronaMalik(I,iter, K, opt_method) {
iter = iter || 10;
K = K || 3;
var method = opt_method || 1;
// Define kernels
var dxW = ee.Kernel.fixed(3, 3,
[[ 0, 0, 0],
[ 1, -1, 0],
[ 0, 0, 0]]);
var dxE = ee.Kernel.fixed(3, 3,
[[ 0, 0, 0],
[ 0, -1, 1],
[ 0, 0, 0]]);
var dyN = ee.Kernel.fixed(3, 3,
[[ 0, 1, 0],
[ 0, -1, 0],
[ 0, 0, 0]]);
var dyS = ee.Kernel.fixed(3, 3,
[[ 0, 0, 0],
[ 0, -1, 0],
[ 0, 1, 0]]);
var lambda = 0.2;
var k1 = ee.Image(-1.0/K);
var k2 = ee.Image(K).multiply(ee.Image(K));
// Convolve
for(var i = 0; i < iter; i++) {
var dI_W = I.convolve(dxW);
var dI_E = I.convolve(dxE);
var dI_N = I.convolve(dyN);
var dI_S = I.convolve(dyS);
// Combine using choosen method
switch(method) {
case 1:
var cW = dI_W.multiply(dI_W).multiply(k1).exp();
var cE = dI_E.multiply(dI_E).multiply(k1).exp();
var cN = dI_N.multiply(dI_N).multiply(k1).exp();
var cS = dI_S.multiply(dI_S).multiply(k1).exp();
I = I.add(ee.Image(lambda).multiply(cN.multiply(dI_N).add(cS.multiply(dI_S)).add(cE.multiply(dI_E)).add(cW.multiply(dI_W))));
break;
case 2:
var cW = ee.Image(1.0).divide(ee.Image(1.0).add(dI_W.multiply(dI_W).divide(k2)));
var cE = ee.Image(1.0).divide(ee.Image(1.0).add(dI_E.multiply(dI_E).divide(k2)));
var cN = ee.Image(1.0).divide(ee.Image(1.0).add(dI_N.multiply(dI_N).divide(k2)));
var cS = ee.Image(1.0).divide(ee.Image(1.0).add(dI_S.multiply(dI_S).divide(k2)));
I = I.add(ee.Image(lambda).multiply(cN.multiply(dI_N).add(cS.multiply(dI_S)).add(cE.multiply(dI_E)).add(cW.multiply(dI_W))));
break;
}
}
return I;
};
//Edge Otsu Algorithms
function edgeOtsu(img,kwargs) {
var geom = img.geometry()
// get list of band names used later
var bandList = img.bandNames();
var kwargKeys = [];
for(var key in kwargDefaults) kwargKeys.push( key );
var params;
var i,k,v;
// loop through the keywords and construct ee.Dictionary from them,
// if the key is defined in the input then pass else use default
params = ee.Dictionary(kwargs);
for (i=0;i<kwargKeys.length;i++) {
k = kwargKeys[i];
v = kwargDefaults[k];
params = ee.Dictionary(
ee.Algorithms.If(params.contains(k),params,params.set(k,v))
);
}
// parameters for all methods
var initialThreshold = ee.Number( params.get('initialThreshold') ),
reductionScale = ee.Number( params.get('reductionScale') ),
smoothing = ee.Number( params.get('smoothing') ),
bandName = ee.String( params.get('bandName') ),
connectedPixels = ee.Number( params.get('connectedPixels') ),
edgeLength = ee.Number( params.get('edgeLength') ),
smoothEdges = ee.Number( params.get('smoothEdges') ),
cannyThreshold = ee.Number( params.get('cannyThreshold') ),
cannySigma = ee.Number( params.get('cannySigma') ),
cannyLt = ee.Number( params.get('cannyLt') ),
maxBuckets = ee.Number( params.get('maxBuckets') ),
minBucketWidth = ee.Number( params.get('minBucketWidth') ),
maxRaw = ee.Number( params.get('maxRaw') ),
invert = params.get('invert'),
verbose = params.get('verbose').getInfo();
// get preliminary water
var binary = img.lt(initialThreshold).rename('binary');
Map.addLayer(binary,{min:0,max:1},"binary threshold")
// get canny edges
var canny = ee.Algorithms.CannyEdgeDetector(binary, cannyThreshold, cannySigma);
Map.addLayer(canny,{},"Canny edge detection")
// process canny edges
var connected = canny.updateMask(canny).lt(cannyLt).connectedPixelCount(connectedPixels, true);
var edges = connected.gte(edgeLength);
edges = edges.updateMask(edges);
Map.addLayer(edges,{},"edges")
var edgeBuffer = edges.focal_max(smoothEdges, 'square', 'meters');
Map.addLayer(edgeBuffer,{},"Edge buffer")
// get histogram for Otsu
var histogram_image = img.updateMask(edgeBuffer);
// histogram_image = histogram_image.clip(geometry2)
var histogram = ee.Dictionary(histogram_image.reduceRegion({
reducer:ee.Reducer.histogram(maxBuckets, minBucketWidth,maxRaw)
.combine('mean', null, true).combine('variance', null,true),
geometry: geom,
scale: reductionScale,
maxPixels: 1e13,
tileScale:16
}).get(bandName.cat('_histogram')));
var threshold = otsu(histogram);
var chart = constructHistChart(histogram,threshold)
.setOptions({
title: 'Edge Search Histogram',
hAxis: {
title: 'Values',
},
vAxis:{
title:'Count'
}
});
print('Algorithm parameters:',params);
print("Calculated threshold:",threshold);
print('Thresholding histogram:',chart);
// segment image and mask 0 values (not water)
var waterImg = ee.Image(ee.Algorithms.If(invert,img.gt(threshold),img.lt(threshold)));
Map.addLayer(waterImg,{palette:"white,blue"},"water image")
return waterImg;
}
function otsu(histogram) {
// make sure histogram is an ee.Dictionary object
histogram = ee.Dictionary(histogram);
// extract relevant values into arrays
var counts = ee.Array(histogram.get('histogram'));
var means = ee.Array(histogram.get('bucketMeans'));
// calculate single statistics over arrays
var size = means.length().get([0]);
var total = counts.reduce(ee.Reducer.sum(), [0]).get([0]);
var sum = means.multiply(counts).reduce(ee.Reducer.sum(), [0]).get([0]);
var mean = sum.divide(total);
// compute between sum of squares, where each mean partitions the data
var indices = ee.List.sequence(1, size);
var bss = indices.map(function(i) {
var aCounts = counts.slice(0, 0, i);
var aCount = aCounts.reduce(ee.Reducer.sum(), [0]).get([0]);
var aMeans = means.slice(0, 0, i);
var aMean = aMeans.multiply(aCounts)
.reduce(ee.Reducer.sum(), [0]).get([0])
.divide(aCount);
var bCount = total.subtract(aCount);
var bMean = sum.subtract(aCount.multiply(aMean)).divide(bCount);
return aCount.multiply(aMean.subtract(mean).pow(2)).add(
bCount.multiply(bMean.subtract(mean).pow(2)));
});
// return the mean value corresponding to the maximum BSS
return means.sort(bss).get([-1]);
}
function constructHistChart(histogram,threshold){
var counts = ee.List(histogram.get('histogram'));
var buckets = ee.List(histogram.get('bucketMeans'));
// construct array for visualization of threshold in chart
var segment = ee.List.repeat(0, counts.size());
var maxFrequency = ee.Number(counts.reduce(ee.Reducer.max()));
var threshIndex = buckets.indexOf(threshold);
segment = segment.set(threshIndex, maxFrequency);
var histChart = ui.Chart.array.values(ee.Array.cat([counts, segment], 1), 0, buckets)
.setSeriesNames(['Values', 'Threshold'])
.setChartType('ColumnChart');
return histChart;
}
//Declare the period
var start = "2020-10-01";
var end = "2020-10-31";
// define default parameterization for keywords
var kwargDefaults = {
'initialThreshold':-14,
'reductionScale': 180,
'smoothing': 100,
'bandName': "VV",
'connectedPixels': 100,
'edgeLength': 20,
'smoothEdges': 20,
'cannyThreshold': 1,
'cannySigma': 1,
'cannyLt': 0.05,
'maxBuckets': 255,
'minBucketWidth': 0.001,
'maxRaw': 1e6,
'invert':false,
'verbose': false
};
// get a few sentinel1 images to run algorithms on
var s1 = ee.ImageCollection("COPERNICUS/S1_GRD")
.filterBounds(geometry)
.filterDate(start,end);
// apply slope correction and speckle filter
s1 = s1.map(slopeCorrection)
.map(PeronaMalik);
var img = ee.Image(s1.first()).select("VV");
Map.addLayer(img,{min:-25,max:0},"s1 image VV");
var otsu = edgeOtsu(img,kwargDefaults);