What we have learned so far
From Profiling
  • -dLsw000.0/000.0/0-1.0
  • Lswgreen=25.9   
  • Lswred=13.9   
  • LswNIR=5.5
 

 
  Activity
  • learn about Extract/Calibrate calibration data
  • nothing to do for now
  • we shall actually do it in the next page
Syntax Extract

Specify the conditions for extraction of calibration data
Now that a Special_Effects_Mask is available in channel_bands+1,
we can specify our own conditions for the extraction of calibration data from the image
  • in #Line_1, we have specified -AutoCAL/NIRmax255/mBPL4/NoSmooth
    • this excludes pixels masked at 1, 2 and 3
    • therefore only allows  pixels masked in the range 4 up to 20
  • this is most unusual, as standard procedure is -AutoCAL/NIRmax255/mBPL2/NoSmooth
    • which only excludes  pixels which have been masked at 1 (if any)
    • and allows all pixels with mSE in the range 2 up to 20
Append #LINE_4 to your script
  • copy it from here
  • and paste/append it in your script
  • Note: the -deglint... argument must be available if deglinting is desired:
    • set it with a lowercase d:               -deglint/....
    • do not disable it with an arobase   @deglint/....
  • Note: the -Extract...   argument is enabled
  • Note: if a -Calibrate... argument is also enabled,
    • first 4SM extracts the calibration data,
      • and write a calibration textfiles tarawa-subset_m0.cal and tarawa-subset_m0.bdh
    • then 4SM proceeds to compute and display the calibration diagram,
      • which is virtually the same as the diagram obtained by the AutoCAL process

choose your pick, be choosy
Syntax Extract
Sample just what you want, ignore the rest, investigate your image

mBPL
  • We know how to take control
    • -Extract/v/mBPL3
      • mBPL3 applies: 
        • would exclude pixels masked at 1 and 2
        • and allow pixels masked at 3 to 20
    • -Extract/v/mBPL4
      • mBPL4 applies:
        • would exclude pixels masked at 1, 2 and 3
        • and allow pixels masked at 4 to 20
    • -Extract/v/mBPL5
      • would yield a void tarawa-subset_m0.cal text file
        • as there is no pixels masked in the range 5 to 20
      • this is how you can sample land only
 

MBPL
  • MBPL
    • using -Extract/v/MBPL3   would only sample pixels masked at 3
      • excluding those masked at 1, 2, 4 to 20
 

 Output is a new  tarawa-subset_m0.cal   text file 
which has overwriten the previous one obtained through the AutoCAL process
 HowTo Read the calibration text file 
  • M_002 1 2................the BPL in the pair_1/2
  • M_002 1 3................the BPL in the pair_1/3
  • M_002 2 3................the BPL in the pair_2/3
  • m_002 1 2................the DPL in the pair_1/2
  • m_002 1 3................the DPL in the pair_1/3
  • m_002 2 3................the DPL in the pair_2/3
  • SL_SOIL...................the average Soil Line under mSE=mSOIL
  • SL_DPL.....................the average Soil Line as an average of all DPL pixels
  • SL_NIR.....................the average Soil Line as an average of all bareland pixels in the pair Red/NIR
  • LogNbPix 1 2.............the BiDim Histogram in the pair_1/2
  • LogNbPix 1 3.............the BiDim Histogram in the pair_1/3
  • LogNbPix 2 3.............the BiDim Histogram in the pair_2/3
     
 
We are ready now for optical calibration under mask_4
next to first Calibration

 
 







 


Things you can do

Look at the location of BPL pixels in the image
This at times helps understanding the calibration of a particular scene
 

pair 12 in channel 15

pair 13 in channel 16

pair 23 in channel 17


Here is a color composite of these three pairs of BPL


 

back to Tarawa tutorial                                                              back to Deglint your image    
next to first Calibration
 


 


 



Créer un site
Créer un site