Image is unbounded Error in Earth Engine Python API after vector to raster conversation by .reduceToImage function

Question

I look at several Sentinel 1 GRD images and want to calculate the median for multiple regions for every image inside the image collection. First, I create an image collection by filtering dates and bounds:

geojsonFc= ee.FeatureCollection({
  "type": "FeatureCollection",
  "features": [
    {
      "type": "Feature",
      "properties": {},
      "geometry": {
        "type": "Polygon",
        "coordinates": [
          [                [
              8.4574556350708,
              49.853355306121884
            ],                [
              8.460803031921387,
              49.853355306121884
            ],                [
              8.460803031921387,
              49.85692454014551
            ],                [
              8.4574556350708,
              49.85692454014551
            ],                [
              8.4574556350708,
              49.853355306121884
            ]]]}}]})



s1_grd = ee.ImageCollection('COPERNICUS/S1_GRD').filterBounds(geojsonFc).filterDate(ee.Date('2018-06-01'), ee.Date('2018-06-30'))

Within the second step, I map a reducer function over the collection

s1_grd_rr = ee.FeatureCollection(s1_grd.map(lambda x: x.reduceRegions(collection=geojsonFc,reducer='median', crs='EPSG:4326',scale=10)))

Now I got a FeatureCollection with several FeatureCollections containing the median of all bands and corresponding geometries and coordinates. I want to transform this information back into a raster(ee.Image). I think the reduceToImage function suits best for this purpose.

Here is an example for the first FeatureCollection from s1_grd_rr

test = ee.FeatureCollection(s1_grd_rr.first()).filter(ee.Filter.notNull(['VV'])).reduceToImage(properties=['VV'], reducer=ee.Reducer.first())

Now I have an ee.Image.

test.getInfo()

{'type': 'Image', 'bands': [{'id': 'first', 'data_type': {'type': 'PixelType', 'precision': 'double'}, 'crs': 'EPSG:4326', 'crs_transform': [1, 0, 0, 0, 1, 0]}]}

Now i want to display these image in Ipython and I try:

import IPython.display as disp
url = test.select('first').getThumbURL({'min': -20, 'max': 0})
disp.Image(url=url, width=800)

But now i get these error:

EEException: Image is unbounded. Must specify a region or a pixel grid with dimensions.

Any ideas what I am doing wrong? I mean, the error say what i have to do. I have to specify a region, but i don't know how to do that ;)

score 0 · Answer 1 · answered Nov 14 '21 at 19:23

Found out that I have to change two lines:

.reproject('epsg:4326',None,10)

 test = ee.FeatureCollection(s1_grd_rr.first()).filter(ee.Filter.notNull(['VV'])).reduceToImage(properties=['VV'], reducer=ee.Reducer.first()).reproject('epsg:4326',None,10)

and .clip(geojsonFc.geometry())

url = test.select('first').clip(geojsonFc.geometry()).getThumbURL({'min': -20, 'max': 0})

with

test.getInfo()

{'type': 'Image', 'bands': [{'id': 'first', 'data_type': {'type': 'PixelType', 'precision': 'double'}, 'dimensions': [497, 252],
'origin': [93924, -555018], 'crs': 'EPSG:4326', 'crs_transform': [8.983152841195215e-05, 0, 0, 0, -8.983152841195215e-05, 0]}]

I see that now a dimension and origin information are added.

Image is unbounded Error in Earth Engine Python API after vector to raster conversation by .reduceToImage function

1 Answers1