GDAL Raster – NoData Value Ignored Issue

gdalrasterstatistics

I am trying to calcualate the statistics of a TIFF. The script should ignore the NoData value, but doesn't. What causes this?

import gdal
from gdalconst import *

# register all of the drivers
gdal.AllRegister()

# open the image
ds = gdal.Open('test_slope.tif', GA_ReadOnly)

# get raster band 1
band = ds.GetRasterBand(1)

# set NoData value
band.SetNoDataValue(-3.4028230607371e+38)

# Calculate and print Statistics
stats = band.GetStatistics(0,1)
print stats

The results look like this

>>> 
[-3.4028230607371e+38, 126.5991897583, -4.6432918375055e+37, 1.1680875238428e+38]
>>>

Best Answer

GetStatistics will reuse previously computed statistics if they exist (i.e computed before you set the NoData value). You can use stats = band.ComputeStatistics(0) instead of GetStatistics to force the statistics to be recomputed.

Related Solutions

[GIS] Create a Raster file out of a ASCII grid file

You can easily open ASCII xyz triplicate data in QGIS under "Add Raster Data" with a "ASCII Gridded XYZ (.xyz)" file type. You can also covert it to a different format under the "Raster > Conversion > Translate (Convert format)" menu. Alternately, you can do this under the "Raster > Conversion > Rasterize" menu with a "Comma Separated Value (.csv)" file type.

Raster – How to Calculate Mean for Each Band Using GDAL

From all the helpful comments and some additional troubleshooting, here is what I found:

QGIS is treating the NaN values (pixels that fall within the extent of the ortho, but are not actually in the ortho - see image) as zeros. So my band averages are darker (closer to zero) then they should be. I sectioned my ortho into multiple pieces and disregarded the sections that contained the edges (for testing purposes) and compared the values from both the original code I posted and from the link user30184 provided to how QGIS calculated their average values and they were negligible.

I also compared the highest res image to a lower res image and found that for my purposes, the mean values were negligible (max difference of around 0.5 units). So for my purpose, I will stick to using the lower res ortho.

Final code:

def rgb_mean(output)
    rgb_df = pandas.DataFrame(columns=['Red', 'Green', 'Blue'])
    raster = gdal.Open(output)
    bands = raster.RasterCount
    avg = []

    for band in range(1, bands + 1):
        data = raster.GetRasterBand(band).ReadAsArray()
        mean = np.mean(data)
        avg.append(mean)

    rgb_df = rgb_df.append({'Red': avg[0], 'Green': avg[1], 'Blue': avg[2]}, ignore_index=True)

I removed the [data != 0] from mean = np.mean(data[data != 0] because I do want to include pixels that may have a zero in their band.

Best Answer

Related Solutions

[GIS] Create a Raster file out of a ASCII grid file

Raster – How to Calculate Mean for Each Band Using GDAL

Related Question