Case Study on Multispectral Land Cover Classification - Open Source Geospatial Tools: Applications in Earth Observation - page 290

(a) (b)

R=B5
G=B4
B=B3

R=B5
G=B4
B=B3

0 25 50 km 0 25 50 km

(c)

R=B5
G=B4
B=B3

0 25 50 km

Fig. 17.2 Individual Landsat 8 scenes and minimum cloud composite a LC82070232013160LG-
N00.tif, b LC82070232013192LGN00.tif, c LC82070232013_composite.tif

-co INTERLEAVE=BAND
Creation option for band sequential encoding (see Sect. 3.5).
-co COMPRESS=LZW
Creation option for LZW compressed output (see Sect. 5.5.4).

To create the cloud free composite in Fig. 17.2 c based on the two multi-band
raster datasets in Fig. 17.2 a, b, we can use the utility pkcomposite from pktools
(see Sect. 12.2 ) .

pkcomposite -i LC82070232013160LGN00.tif -i
ₒ

LC82070232013192LGN00.tif -o LC82070232013_composite.tif

ₒ

-bndnodata 8 -srcnodata 1 -dstnodata 0 -cr maxndvi -cb 3

ₒ

-cb 4