问题
i have a geotiff images that have 3bands.
band1,2 is a actual image values and band3 is a instance angle value.
band1,2 is float32 data type
under code is that i try before.
but it doesn't work.
i think band data's range is too large, so it doesn't
from osgeo import gdal, osr, ogr
from PIL import Image
import numpy as np
ds = gdal.Open('image path', gdal.GA_ReadOnly)
rb = ds.GetRasterBand(1)
test = rb.ReadAsArray()
rb2 = ds.GetRasterBand(2)
test2 = rb2.ReadAsArray()
rb3 = ds.GetRasterBand(3)
test3 = rb3.ReadAsArray()
slice56 = test2
formatted = (slice56 * 255 / np.max(slice56)).astype('uint8')
img = Image.fromarray(formatted)
img.save('save image path')
how can i solve this problem??
回答1:
You can use gdal.Translate
for this.
You can read the documentation here
from osgeo import gdal
options_list = [
'-ot Byte',
'-of JPEG',
'-b 1,
'-scale'
]
options_string = " ".join(options_list)
gdal.Translate('save_image_path.jpg',
'image_path.tif',
options=options_string)
The above code simply create a jpg file with band 1 scaled into byte range. You could add more bands by adding, '-b 2'
etc. Also notice that scale automatically wraps the entire range into byte range. If you like something else you could use '-scale min_val max_val'
in order to specify the range you like, since often you have no need of either the lowest or highest values available.
回答2:
The above worked well for me except the JPEG resolution wasn't great. Swapping JPEG to PNG worked better.
来源:https://stackoverflow.com/questions/50207292/how-to-convert-geotiff-to-jpg-in-python-or-java