User Tools

Site Tools


Basic GDAL and OGR script

Read and Manipulate image metadata

Script: open by kate ~/ost4sem/exercise/basic_adv_gdalogr/
Directory: ~/ost4sem/exercise/basic_adv_gdalogr

gedit ~/ost4sem/exercise/basic_adv_gdalogr/ &

Change the directory, see the image input.tif with openev, run gdalinfo and return to the questions:

cd ~/ost4sem/exercise/basic_adv_gdalogr
openev input.tif
gdalinfo input.tif

What is the pixel size?
How do you get min and max pixel values?
The image has no information on a projection.
From the pixel values and from the boundary I can understand that it is in lat-long degree (WGS84).
Let's apply a projection using gdal_warp

gdal_translate -co COMPRESS=DEFLATE -co ZLEVEL=9 -a_srs EPSG:4326  input.tif input_proj.tif
# or you can also use gdalwarp
gdalwarp -co COMPRESS=DEFLATE -co ZLEVEL=9  -t_srs EPSG:4326  -s_srs EPSG:4326  input.tif input_proj.tif
gdalinfo input_proj.tif

Calculate minimum and maximum values for all the images:

cd ~/ost4sem/exercise/basic_adv_gdalogr/fagus_sylvatica
for file in *.tif ; do 
gdalinfo  -mm  $file | grep "Computed Min/Max" 

Use one of the gdal commands and try to clip all the images using a square box with lat &lon that you like.

  1. Open gedit and save the the script with the extension .sh
  2. Create a working directory following this procedure
  3. Copy the ~/ost4sem/exercise/basic_adv_gdalogr/fagus_sylvatica/*.tif in the $INDIR
  4. Identify the gdal command going to the gdal internet site
  5. Identify the option of the command to make the clip.
  6. Identify the square box coordinates using openev.
  7. Do a test with one image. Open it by openev.
  8. Create a loop for all the images.

Use all the previous examples stored in this wiki. All the commands have already been explained - you just need to combine them.
Use the $OUTDIR $INDIR varibles to define the directories of the files. Write comments and explanation of the commands. Your script should be able to be run the script by sh

Understand data type

                      Ranges of GDAL data types        Image
GDAL data type	       minimum  	maximum        Size 
Byte  			     0  	    255        39M
UInt16			     0  	 65,535        78M
Int16, CInt16 	       -32,768  	 32,767        78M
UInt32			     0    4,294,967,295        155M  
Int32, CInt32 	-2,147,483,648    2,147,483,647        155M
Float32, CFloat32      -3.4E38           3.4E38        155M
Float64, CFloat64    -1.79E308         1.79E308        309M

From Image to text and from txt to image

cd ~/ost4sem/exercise/basic_adv_gdalogr
gdal_translate -of  XYZ   input.tif  output.txt
awk '{if ($3>0.01 && $3<0.9) {print $1,$2,50 } else {print}}' output.txt > output_change.txt
gdal_translate output_change.txt output_change.tif 

The use of VRT to stack images and Tiling

gdalbuildvrt -overwrite -separate   stack.vrt   20?0_TSSP_IP-ENS-A2-SP20_43023435.tif
gdal_translate  -srcwin 0       0 1670 2015   stack.vrt     stack_UL.tif
gdal_translate  -srcwin 0    2015 1670 2015   stack.vrt     stack_UR.tif
gdal_translate  -srcwin 1670    0 1670 2015   stack.vrt     stack_LL.tif
gdal_translate  -srcwin 1670 2015 1670 2015   stack.vrt     stack_LR.tif

Use a stack vrt to build up tailed VRT

gdalbuildvrt -overwrite -te  4302000 3435000 5972000 5450000   stack_LL.vrt  stack.vrt

Recompose the image

gdalbuildvrt -overwrite mosaic.vrt stack_UL.tif stack_UR.tif stack_LL.tif stack_LR.tif
gdal_translate mosaic.vrt mosaic.tif

Merge Images

Merge different ASTER dem and change the projection.

# merge the tiles  -o ASTGTM_merge/ASTGTM_dem_wgs84.tif  */ASTGTM_*_dem.tif
# change the projection
gdalwarp -co "COMPRESS=LZW" -t_srs "+proj=utm +zone=36 +south +a=6378249.145 +b=6356514.870 +units=m +towgs84=-160,-6,-302" ASTGTM_merge/ASTGTM_dem_wgs84.tif ASTGTM_merge/tmp.tif
# compress the file 
gdal_translate -co "COMPRESS=LZW"   ASTGTM_merge/tmp.tif  ASTGTM_merge/ASTGTM_dem_tz36.tif 
# remove intermediate file
rm ASTGTM_merge/tmp.tif ASTGTM_merge/ASTGTM_dem_wgs84.tif

open it by openev

 openev ASTGTM_merge/ASTGTM_dem_tz36.tif

Vector manipulation with OGR

ogrinfo programme
ogrinfo -al
ogrinfo -al -geom=NO    poly_fr_10poly.shp

# select a field 

ogrinfo -al -geom=NO    poly_fr_10poly.shp | grep id
Look at a different file
cd ~/ost4sem/exercise/KenyaGIS
ogrinfo -al County.shp
ogrinfo -al County.shp | head -30
ogrinfo -al County.shp | head -41
Process vector file with ogr2ogr and ogrinfo
ogr2ogr  -f "GML" -s_srs EPSG:4326  -t_srs EPSG:3857 -select COUNTY  -where "COUNTY = 'Isiolo'"  /tmp/Isiolo.gml  $INPUT/County.shp
# Extract bounding box for different county and overall county in Kenya
ogrinfo -al County.shp | grep "Extent" | awk '{gsub(/\(/,"");gsub(/\)/,"");gsub(/,/,"");print $2" "$3" "$5" "$6} '
ogrinfo -al /tmp/Isiolo.gml | grep "Extent" | awk '{gsub(/\(/,"");gsub(/\)/,"");gsub(/,/,"");print $2" "$3" "$5" "$6} '
wiki/gdalbasic.txt · Last modified: 2017/06/20 07:21 by ost4sem_ste