Optical calibration, bathymetry, water column correction and bottom typing of shallow marine areas, using passive remote sensing imageries
WorldView 2 image at Oahu, Hawaii islands
3289x3241, 2 m ground resolution, courtesy of Ron Abileah
[back to wainanalo beach]


Deglinting
next you're in for a big surprise: Deglinting a "deglinted" image


 
1 - NO NEED for field data, nor for atmospheric correction
2 - this is demonstrated in this website, using a variety of hyper/multi spectral data
 
Requirements are
1 - homogeneous water body and atmosphere
2 - some coverage of optically deep water
3 - some coverage of dry land
 
Problems are
1 - the precision on estimated depth is found wanting, because the noise-equivalent change in radiance  of accessible data is too high for shallow water column correction work 
2 - radiance data should be preprocessed by the provider at level 1 in order to improve S/N ratio
3 - exponential decay: the deeper/darker the bottom, the poorer the performances
 
So
I keep digging
until suitable data
become available
 


Imaging geometry
from 11MAR31215109-M2AS-052656310070_01_P002.IMD from DigitalGlobe
looking right into the sun and sideways!!

    meanSunAz = 146.5;
    meanSunEl =  70.0;
    meanSatAz = 306.1; sensor looks SE, right into the sun
    meanSatEl =  47.1;
    meanInTrackViewAngle =  14.9;
    meanCrossTrackViewAngle = -34.8;
    meanOffNadirViewAngle =  37.5; sensor looks full sideways


April 10th
Glint Correlations


Glint regressions based on band 8
  • Again, it would seem that this image is not raw from data provider: "the band images after subtracting an estimate of the deep water upwelling radiance"??

  • Pixels at Zero cause a bad distortion of the glint regression: they MUST be bypassed

 

 
Glint regressions based on band 7
 

BiDim histograms over deep waters
band 8 is well correlated with bands 6, 4 and 1

BiDim histograms over deep waters
band 7 is well correlated with bands 5, 3 and 2

  
7 vs 8                            6 vs 8

  
3 vs 8                            2 vs 8
 
  
5 vs 8                            4 vs 8

  1 vs 8
  
  6 vs 7                           5 vs 7

  
2 vs 7                            1 vs 7
  
  4 vs 7                           3 vs 7


 
A weird pattern of interband correlations is observed:
  • NIR band 8 is well correlated with bands 6, 4 and 1
  • NIR band 7 is well correlated with bands 5 , 3 and 2
  • this must be accounted for to improve deglinting
 
 

  • Glint in bands 7 and 8 are offset
  • Lowest radiance goes down to zero in raw data
    • I had to offset them a bit, as ZERO is NODATA by default!
 

bands 2, 3, 5 and 7
  • Note the residues of glint in deglinted data in spite of plain strong smoothing and of smart deglinting

 

bands 1, 4, 6 and 8
  • See that the glint regressions yield suitable estimation of spectral deep water radiances:
  • -Lsw/044.7/029.3/010.6/009.3/009.5/000.0/000.1/000.1




April 10th
Deglinting


Raw color composite 
for bands 2, 4 and 4

4SM Deglinted color composite 
for bands 2, 4 and 4

TCC raw data

4SM TCC deglinted data
 
   

next you're in for a big surprise: Deglinting a "deglinted" image