3

I have the satellite data files (2) which are available as satdat and satdat.hdr. As per the answer provided here, I tried the following, but, it gave me an invalid header format error.

import imageio
from pathlib import Path

imageio.plugins.freeimage.download()

datapath = Path(r'./sat_data/')

filename = str(datapath / 'satdat.hdr')    # I even tried not using this extension
im = imageio.imread(filename, format='HDR-FI')
print(im.shape)

I was expecting this to read the data into numpy arrays; rather it threw the error as invalid ENVI header file. Looking forward to any library to open this type of file, except OpenCV (due to certain constraints on the compute environment I am using.)

anurag
  • 1,715
  • 1
  • 8
  • 28

1 Answers1

3

After much trial and error, the following worked for me:

import spectral.io.envi as envi
from pathlib import Path

datapath = Path(r'./sat_data/')

header_file = str(datapath / 'satdat.hdr')
spectral_file = str(datapath / 'satdat')

numpy_ndarr = envi.open(header_file, spectral_file)

img = numpy_ndarr.read_bands([10, 11, 12])  # select the bands

# Here, img is an numpy nd-array

print(img.shape)    # returns (100, 100, 3)

I hope others may find this useful. The spectral library can be found here

anurag
  • 1,715
  • 1
  • 8
  • 28