Calibrate the model parameters
First calibration under red mask

to Tarawa tutorial       
to Prepare your image
to Deglint/investigate your image 
to Extract calibration data
to Calibrate under Mask=4
to Model      under Mask=4
to Calibrate  under Mask=3
to Model the whole lagoon

 
 
 
Things 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 how to take control of the -Calibrate... command line argument
  • nothing to do for now

First calibration    under RED mask_4

We are happy with our special effect mask
We have our two calibration data text files

  • tarawa-subset_m0.cal ..........for BPL and DPL pixels
  • tarawa-subset_m0.bdh..........for bi-dimensional histograms

We are ready to specify the following model parameters

  • the Brightest Pixels Line (BPL)
  • the Soil Line (SL)
  • the spectral BOA radiances LM and Lw
  • the complete series of Ki/Kj ratios
  • spectral K values for all visible bands

Syntax     Calibrate         KK         LsM         Lsw         Lw         WL         CP

Howto         Calibrate         Calibration TextFile         Calibration Diagram
 

CALIBRATION : various arguments

-WL... argument
Wavelengths must be adjusted carefully, as they are key to Jerlov's water types.

For SPOT images, we have established
that WL[XS1]=550, WL[XS2]=650 and WL[XS3]=850 nm are convenient.
these wavelengths are imbedded in 4SM for SPOT images:
they need not to appear in the commandline.
-LsM...    -LsM/199.5/179.8/166.8_AutoCAL
  • LsM is the spectral radiance
    • at the intersection of the Brightest Pixels Line with the Soil Line.
-Lsw...          -Lsw/026.6/013.6/006.5_AutoCAL
 
-dLsw...      -dLsw001.0/00.6/00.0
  • provides an adjustment to Lsw values: Lsw=Lsw+dlsw
-Lw... -Lw/0001.5/000.0/000.0_AutoCAL
 
-CP...   -CP/160.50/102.06/014.11_0.75m
a spectral calibration pixel. 
  • If a -CP argument is needed, it may be chosen from the calibration screen display
  • This -CP... initial spectral value 
    • is then progressively adjusted manually, along with the -LsM spectral value,
    • so as to obtain a good fit of the BPL model onto the observed BPL pixel.
  • In the end, the BPL model 
    • shall be the straight line after linearization of the data
    • this staright line runs through the calibration pixel and the LsM point , 
    • and its slope is the ratio Kgreen/Kred. 
-KK....         -KK1_2_0.29/Knir4.500/mask_3
states that

  • the ratio K1/K2..........is set at 0.29  for the pair 1/2

-Z....          -Z/MSL0.00/n_0/mask_3

  • specifies some parameters for the computation of depth for each BPL pixel.
-B.....    -B/tclNe/Bmin0/cLM1.00
  • specifies conditions relative to brightness of pixels.
 -deglint...        -deglint/vRbaD/GlintM25.0
  • because the data is affected by sky/sun glint, this argument specifies conditions for deglinting.

-extract...       -extract/v/mBPL4/mask_3

  • provides details of conditions applied upon extraction of the calibration data textfile.
  • this argument is mandatoty for the calibration process
    • ?may only be disabled with a lowercase -extract
-Calibrate...  -Calibrate/v/BdSNpZg/BDh_12_13
  • NOTE: the -deglin/t..... and -extract/.... arguments 
    • may/should be available
    • for the -Calibrate... process: do not disable them
  • NOTE: if the calibration data file is not available,
    • 4SM shall first extract the calibration data before proceeding to the calibration process
Calibration : RunTime
  • Upon execution of the calibration command line 
    • the output is an executable GMT script text file named calibration.sh
    • calibration.sh is then executed automatically to produce and display a PostScript graphic calibration diagram on screen
    • tarawa-subset_calibration_Line_4: your aim is to fine-tune this plot,
      • step by step
      • starting from this first plot,
      • making some adjustment to one parameter at a time
      • runing the command line again
      • making some more adjustment
      • etc
      • eventually modifying the conditions for extraction of calibration data
      • etc
      • until satisfied that the model parameters are consistent with
        • the basics of underwater and atmospheric optics
        • the simplifying assumptions behind the 4SM process
        • the realities of the radiances that have been captured by the remote sensor.
    • "gv -scale 0.6667 -landscape tarawa-subset_3_2_1.eps & "
  • By close inspection of this display ,
    • the game consists in gradually and iteratively optimizing -KK, -LsM, -Lsw, and -Lw arguments in the command line,
    • until a satisfactory operational optical model is established.
  • This iterative process is at the heart of 4SM optical calibration.
 

         
LINE_4
Gradually tune your calibration

Activity  : use LINE_4

  • Calibrate the model under mask 4
    • let's do it, at last!

HowTo KK
Iteration on bands to process
Plot options
Z2 vs Z3
Update your script
LINE 4
LINE 4: calibrate under mask 4
please, append a copy of #LINE_4 into your script
then, do as directed in the list of your actions, step by step


 		 <<==A nice calibration, under mask 4
 
But we are in for a big surprise!

Depth in centimeters
Location of Profile_2 is shown

Computed depth along Profile_2
 
back to Extract calibration data

next to Calibrate page 2
   



Créer un site
Créer un site