How to extract elevation using contours in Google Earth Engine?

Question

I want to extract the elevation values and generate contour lines from SRTM DEM at a 100 m intervals. However, I was able to create the contour lines using a code available here. But I am not able to interpret the exported result (shapefile) format. The attribute of the exported result is shown below.

I am not able to understand, what the field "count" mean. Also, I want to extract the elevation values corresponding to each contour line. Is it possible to extract the elevation values in Google Earth Engine using the generated contours?

I am stuck after this much progress.

var srtm = ee.Image('USGS/SRTMGL1_003');

var lines = ee.List.sequence(0, 10000, 100)
 
var contourlines = lines.map(function(line) {
var mycontour = srtm
.convolve(ee.Kernel.gaussian(5, 3))
.subtract(ee.Image.constant(line)).zeroCrossing()
.multiply(ee.Image.constant(line)).toFloat()
.clip(table)
 
return mycontour.mask(mycontour);
})
 
contourlines = ee.ImageCollection(contourlines).mosaic()
 
Map.addLayer(contourlines, {min: 0, max: 5000, palette:['00ff00', 'ff0000']}, 'contours')
Map.centerObject(table,5)
var contour = contourlines.toInt();
var vectors = contour.reduceToVectors({
  geometry: table, 
  scale: 30,
  maxPixels: 1e13
})


Export.table.toDrive({
  collection: vectors,
  description: "ContourLines",
  folder:"DEM",
  //geometry: NE,
  fileFormat: "SHP",
})

'table' is the extent of my study area.

Answer 1

Your code works as intended, but it is returning more information than you want. The function ee.Image.reduceToVectors does two things. First, it looks for homogeneous regions within an integer image, which should be the first band in the input ee.Image:

The first band is expected to be an integer type; adjacent pixels will be in the same segment if they have the same value in this band.

Second, it applies a reducer to all the other bands within the image. The default reducer is ee.Reducer.countEvery. The documentation for this reducer is lacking.. with the only description being:

Returns a Reducer that computes the number of inputs.

What this means for the reduceToVectors operation is that it counts the number of pixels (at the scale you specified, e.g. 30 m) within each polygon. For example, the following feature (obtained using your code above) shows a count value of 16:

This is actually a good default value for the reducer. The "label" corresponds to the value from the first band, i.e. the value of the "homogeneous region". In your case, it actually corresponds to the elevation value, in integer format. The DEM you used seems to actually come in integer format anyway (although it is presented as float). Combining this with the fact that you are generating the contours at 100m of elevation steps, in this specific example the labels correspond exactly to the elevation value you want.

In other cases, you might want to use a different reducer to gather information about other bands in the image. That way, you will get the "label" property (based on the value of the first band, which must be integer), and other properties depending on the other bands and which reducer you use.