photomet

This program performs photometric corrections on a cube. It can be used to normalize an image for mosaicking. The user must input the cube with a pvl containing the corresponding photometric parameters. Examples of sample pvl files are shown below. There are three aspects of photometric correction that need to be considered when running the photomet program:

    1) Surface photometric function model type
    2) Atmospheric photometric function model type
    3) Type of normalization to be performed
    

The types of surface photometric function model types currently available in photomet are:

    HapkeHen        : Hapke-Henyey-Greenstein photometric model. Derive model 
                      albedo using complete Hapke model with Henyey-Greenstein 
		      single-particle phase function whose coefficients are 
		      Hg1 and Hg2, plus single scattering albedo Wh, opposition
		      surge parameters Hh and B0, and macroscopic roughness
		      Theta.
    Lambert         : Simple photometric model which predicts that light
                      incident on a surface is scattered uniformly in all
		      directions; the total amount of reflected light depends
		      on the incidence angle of the illumination. This function
		      does not depend upon the outgoing light direction.
    LommelSeeliger  : This model takes into account the radiance that results
                      from single scattering (scattering of collimated
		      incident light) and does not take into account the
		      radiance that results from multiple scattering (scattering
		      of diffuse light which has made its way indirectly to
		      the same position by being scattered one or more times).
		      This model depends on the incidence and emission angles.
    LunarLambert    : This model combines a weighted sum of the LommelSeeliger
                      and Lambert models. Given a suitable value for the
		      LunarLambert function weight, L, this model fits the true
		      reflectance behavior of many planetary surfaces equally
		      well as the Hapke model. This model also depends on the
		      incidence and emission angles.
    Minnaert        : This model expands upon the Lambert function by introducing
                      constant, K, that is used to describe the roughness of a
		      surface. When the K constant is set to 1.0, then the 
		      Minnaert model is equivalent to the Lambert model.
    LunarLambertMcEwen : This model was developed specifically for use with the
                      Moon. This model was designed to be used in conjunction
		      with the MoonAlbedo normalization model.
    

The types of atmospheric photometric function model types currently available in photomet are:

    Anisotropic1    : Uses Chandrasekhar's solution for anisotropic scattering
                      described by a one term Legendre polynomial. This model
		      uses first order scattering approximation.
    Anisotropic2    : Uses Chandrasekhar's solution for anisotropic scattering
                      described by a one term Legendre polynomial. This model
		      uses second order scattering approximation. It is slower
		      but more accurate than Anisotropic1.
    HapkeAtm1       : Provides an approximation for strongly anisotropic
                      scattering that is similar to Hapke's model for a planetary
		      surface. The Chandrasekhar solution for isotropic
		      scattering is used for the multiple scattering terms, and
		      a correction is made to the singly scattered light for
		      anisotropic particle phase function. A one term Henyey
		      Greenstein function is used. This model uses a first
		      order scattering approximation.
    HapkeAtm2       : Provides an approximation for strongly anisotropic
                      scattering that is similar to Hapke's model for a planetary
		      surface. The Chandrasekhar solution for isotropic
		      scattering is used for the multiple scattering terms, and
		      a correction is made to the singly scattered light for
		      anisotropic particle phase function. A one term Henyey
		      Greenstein function is used. This model uses a second
		      order scattering approximation. It is slower but more
		      accurate than HapkeAtm1.
    Isotropic1      : Uses Chandrasekhar's solution for isotropic scattering.
                      This model uses first order scattering approximation.
    Isotropic2      : Uses Chandrasekhar's solution for isotropic scattering.
                      This model uses second order scattering approximation.
		      It is slower but more accurate than Isotropic1.
    

The types of normalization models currently available in photomet are:

    Albedo          : Normalization without atmosphere. Each pixel is divided by 
                      the model photometric function evaluated at the geometry
                      of that pixel, then multiplied by the function at reference
		      geometry with incidence and phase angles equal to Incref
		      and emission angle 0. This has the effect of removing
		      brightness variations due to incidence angle and showing
		      relative albedo variations with the same contrast
		      everywhere. If topographic shading is present, it will be
		      amplified more in regions of low incidence angle and will
		      not appear uniform.
    AlbedoAtm       : Normalization with atmosphere. For each pixel, a model of 
                      atmospheric scattering is subtracted and a surface model
                      is divided out, both evaluated at the actual geometry of
		      the pixel. Then the resulting value is multiplied by the 
		      surface function at reference conditions is added. In 
		      normal usage, the reference condition has normal incidence
		      (Incref=0) and no atmosphere (Tauref=0) but in some cases
		      it may be desirable to normalize images to a different
		      incidence angle or a finite optical depth to obtain a more
		      uniform appearance. As with the Albedo model, if 
		      topographic shading is present, it will be amplified more at
		      high incidence angles and will not appear uniform.
    Mixed           : Normalization without atmosphere. Used to do albedo 
                      normalization over most of the planet, but near the 
                      terminator it will normalize topographic contrast to avoid
		      the seams that can occur with the usual albedo normalization.
		      The two effects will be joined seamlessly at incidence 
		      angle Incmat. Incmat must be adjusted to give the best
		      equalization of contrast at all incidence angles. The
		      Albedo parameter must also be adjusted so the topographically
		      normalized regions at high incidence angle are set to an
		      albedo compatible with the albedo-normalized data at lower
		      incidence.
    MoonAlbedo      : Normalization without atmosphere. This model was designed
                      specifically for use on Lunar data. It will compute 
                      normalized albedo for the Moon, normalized to 0 degrees
		      emission angle and 30 degrees illumination and phase angles.
		      The LunarLambertMcEwen photometric function was designed to 
                      be used with this normalization model.
    NoNormalization : Normalization without atmosphere. No normalization is 
                      performed. Only photometric correction is performed.
    Shade           : Normalization without atmosphere. The surface photometric 
                      function is evaluated at the geometry of the image in order
                      to calculate a shaded relief image of the ellipsoid (and
		      in the future the DEM). The radiance of the model surface
		      is set to Albedo at incidence angle Incref and zero phase.
		      The image data is not used.
    ShadeAtm        : Normalization with atmosphere. The surface photometric 
                      function is used to simulate an image by relief shading,
                      just like the Shade model, but the effects of atmospheric
		      scattering are also included in the calculation.
    Topo            : Normalization without atmosphere. Used to normalize 
                      topographic shading to uniform contrast regardless of 
                      incidence angle. Such a normalization would exagerate
		      albedo variations at large incidence angles, so this model
		      is used as part of a three step process in which (1) the
		      image is temporarily normalized for albedo; (2) a highpass
		      divide filter is used to remove regional albedo variations;
		      and (3) the image is renormalized with the Topo mode to 
		      undo the first normalization and equalize topographic 
		      shading. The reference state in the first step MUST have
		      Incref=0 because this is waht is undone in the final step.
		      If there are no significant albedo variations, step (2)
		      can be skipped but step (1) must not be.
    TopoAtm         : Normalization with atmosphere. As with the Topo model, 
                      this option is used in the final step of a three step
		      process: (1) normalize with the AlbedoAtm model, Incref=0,
		      and Tauref=0 to temporarily remove atmosphere and normalize
		      albedo variations; (2) use highpass divide filter to remove
		      albedo variations; and (3) normalize with the TopoAtm model
		      to undo the temporary normalization and equalize topographic
		      shading.
    

As you can see above, the only normalization models that make use of atmospheric correction are: AlbedoAtm, ShadeAtm, and TopoAtm. Atmospheric correction is not applied by any of the other normalization models. If you specify an atmospheric model in a PVL along with a normalization model that does not do atmospheric correction, then the atmospheric model will be ignored. Each of the above photometric, atmospheric, and normalization models has specific parameters that apply to them. Here is a list of the models and their related parameters (in parentheses):

    HapkeHen (B0,Hg1,Hg2,Hh,Theta,Wh)
    Lambert
    LommelSeeliger
    LunarLambert (L)
    Minnaert (K)
    LunarLambertMcEwen
  
    Anisotropic1 (Bha,Bharef,Hnorm,Nulneg,Tau,Tauref,Wha,Wharef)
    Anisotropic2 (Bha,Bharef,Hnorm,Nulneg,Tau,Tauref,Wha,Wharef)
    HapkeAtm1 (Hga,Hgaref,Hnorm,Nulneg,Tau,Tauref,Wha,Wharef)
    HapkeAtm2 (Hga,Hgaref,Hnorm,Nulneg,Tau,Tauref,Wha,Wharef)
    Isotropic1 (Hnorm,Nulneg,Tau,Tauref,Wha,Wharef)
    Isotropic2 (Hnorm,Nulneg,Tau,Tauref,Wha,Wharef)
  
    Albedo (Albedo,Incmat,Incref,Thresh)
    AlbedoAtm (Incref)
    Mixed (Albedo,Incmat,Incref,Thresh)
    MoonAlbedo (Bsh1,D,E,F,G2,H,Wl,Xb1,Xb2,Xmul)
    NoNormalization
    Shade (Albedo,Incref)
    ShadeAtm (Albedo,Incref)
    Topo (Albedo,Incref,Thresh)
    TopoAtm (Albedo,Incref)
    

Here is a description of each parameter along with a valid range of values and the default for that parameter:

    Photometric parameters:
    -----------------------
    B0:     Hapke opposition surge component: 0 <= value : default is 0.0
    Bh:     Hapke Legendre coefficient for single particle phase function:
            -1 <= value <= 1 : default is 0.0
    Ch:     Hapke Legendre coefficient for single particle phase function:
            -1 <= value <= 1 : default is 0.0
    Hg1:    Hapke Henyey Greenstein coefficient for single particle phase
            function: -1 < value < 1 : default is 0.0
    Hg2:    Hapke Henyey Greenstein coefficient for single particle phase
            function: 0 <= value <= 1 : default is 0.0
    Hh:     Hapke opposition surge component: 0 <= value : default is 0.0
    K:      Minnaert function exponent: 0 <= value : default is 1.0
    L:      Lunar-Lambert function weight: no limit : default is 1.0
    Theta:  Hapke macroscopic roughness component: 0 <= value <= 90 :
            default is 0.0
    Wh:     Hapke single scattering albedo component: 0 < value <= 1 : 
            default is 0.5
  
    Atmospheric parameters:
    -----------------------
    Bha    : Coefficient of the single particle Legendre phase function:
             -1 <= value <= 1 : default is 0.85
    Hga    : Coefficient of single particle Henyey Greenstein phase
             function: -1 < value < 1 : default is 0.68
    Hnorm  : Atmospheric shell thickness normalized to the planet radius:
             0 <= value : default is .003
    Nulneg : Determines if negative values after removal of atmospheric
             effects will be set to NULL: YES or NO : default is NO
    Tau    : Normal optical depth of the atmosphere: 0 <= value : default 
             is 0.28
    Tauref : Reference value of Tau to which the image will be
             normalized: 0 <= value : default is 0.0
    Wha    : Single scattering albedo of atmospheric particles: 
             0 < value < 1 : default is 0.95
  
    Normalization parameters:
    -------------------------
    Albedo : Albedo to which the image will be normalized: no limit : 
             default is 1.0
    Bsh1   : Albedo dependent phase function normalization parameter: 
             0 <= value : default is 0.08
    D      : Albedo dependent phase function normalization parameter: no 
             limit : default is 0.14
    E      : Albedo dependent phase function normalization parameter: no 
             limit : default is -0.4179
    F      : Albedo dependent phase function normalization parameter: no 
             limit : default is 0.55
    G2     : Albedo dependent phase function normalization parameter: no 
             limit : default is 0.02
    H      : Albedo dependent phase function normalization parameter: no 
             limit : default is 0.048
    Incmat : Specifies incidence angle where albedo normalization transitions 
             to incidence normalization: 0 <= value < 90 : default is 0.0
    Incref : Reference incidence angle to which the image will be normalized:
             0 <= value < 90 : default is 0.0
    Thresh : Sets upper limit on amount of amplification in regions of small
             incidence angle: no limit : default is 30.0
    Wl     : Wavelength in micrometers of the image being normalized: no 
             limit : default is 1.0
    Xb1    : Albedo dependent phase function normalization parameter: no 
             limit : default is -0.0817
    Xb2    : Albedo dependent phase function normalization parameter: no 
             limit : default is 0.0081
    Xmul   : Used to convert radiance to reflectance or apply a calibration 
             fudge factor: no limit : default is 1.0
    

Here are some example PVL files:

    Example 1:
  
    Object = PhotometricModel
      Group = Algorithm
        Name = Lambert
      EndGroup
    EndObject
    Object = NormalizationModel
      Group = Algorithm
        Name = NoNormalization
      EndGroup
    EndObject
  
    --------------------------------
    Example 2:
  
    Object = PhotometricModel
      Group = Algorithm
        Name = Minnaert
	K = .5
      EndGroup
    EndObject
    Object = NormalizationModel
      Group = Algorithm
        Name = Albedo
	Incref = 0.0
	Incmat = 0.0
	Albedo = 1.0
	Thresh = 30.0
      EndGroup
    EndObject
  
    --------------------------------
    Example 3:
  
    Object = PhotometricModel
      Group  = Algorithm
        Name = HapkeHen
	Wh = 0.52
	Hh = 0.0
	B0 = 0.0
	Theta = 30.0
	Hg1 = .213
	Hg2 = 1.0
      EndGroup
    EndObject
    Object = AtmosphericModel
      Group = Algorithm
        Name = HapkeAtm2
	Hnorm = .003
	Tau = 0.28
	Tauref = 0.0
	Wha = .95
	Hga = 0.68
      EndGroup
    EndObject
    Object = NormalizationModel
      Group = Algorithm
        Name = AlbedoAtm
	Incref = 0.0
      EndGroup
    EndObject

    --------------------------------
    Example 4 (Used to process Clementine UVVIS filter "a" data):
  
    Object = PhotometricModel
      Group  = Algorithm
        Name = LunarLambertMcEwen
      EndGroup
    EndObject
    Object = NormalizationModel
      Group = Algorithm
        Name = MoonAlbedo
	D = 0.0
	E = -0.222
	F = 0.5
	G2 = 0.39
	H = 0.062
	Bsh1 = 2.31
      EndGroup
    EndObject
  
    --------------------------------
    Example 5 (Used to process Clementine UVVIS filter "b" data):
  
    Object = PhotometricModel
      Group  = Algorithm
        Name = LunarLambertMcEwen
      EndGroup
    EndObject
    Object = NormalizationModel
      Group = Algorithm
        Name = MoonAlbedo
	D = 0.0
	E = -0.218
	F = 0.5
	G2 = 0.4
	H = 0.054
	Bsh1 = 1.6
      EndGroup
    EndObject
  
    --------------------------------
    Example 6 (Used to process Clementine UVVIS filter "cde" data):
  
    Object = PhotometricModel
      Group  = Algorithm
        Name = LunarLambertMcEwen
      EndGroup
    EndObject
    Object = NormalizationModel
      Group = Algorithm
        Name = MoonAlbedo
	D = 0.0
	E = -0.226
	F = 0.5
	G2 = 0.36
	H = 0.052
	Bsh1 = 1.35
      EndGroup
    EndObject
  
  

This example shows the photometric correction of a cube by using photomet application with the default parameter values kept for the maximum emission and incidence angles.

photomet photomet from=input.cub phopar=input.pvl to=output.cub

photomet GUI using defaults

Example GUI Screenshot of GUI version of the application. Notice the default values of 90.0 are left alone for the MAXEMISSION and MAXINCIDENCE parameters.

Input image

Example input cube Parameter Name: FROM Screenshot of the input image before photometric correction has been performed.

Example input PVL	Pvl file including the input photometric parameters for the input cube.

Output image

Example output cube Parameter Name: TO Screenshot of the output image after photometric correction. Notice many features on the surface are easier to view due to the normalization of the pixels.

This example shows the photometric correction of a cube by using photomet application with the maximum emission angle set to 75.0 and the default parameter value kept for the maximum incidence angle.

photomet photomet from=input.cub phopar=input.pvl to=output.cub maxemission=75.0

photomet GUI with MAXEMISSION changed

Example GUI Screenshot of GUI version of the application. Notice the MAXEMISSION parameter is changed from the default value of 90.0.

Input image

Example input cube Parameter Name: FROM Screenshot of the input image before photometric correction has been performed.

Example input PVL	Pvl file including the input photometric parameters for the input cube.

Output image

Example output cube Parameter Name: TO Screenshot of the output image after photometric correction. Notice there are more blackened areas than in the Example 1 output image. This happens since the pixel values have been set to null where the emission angles are greater than 75.

This example shows the photometric correction of a cube by using photomet application with the maximum incidence angle set to 85.0 and the default parameter value kept for the maximum emission angle.

photomet photomet from=input.cub phopar=input.pvl to=output.cub maxincidence=85.0

photomet GUI with MAXINCIDENCE changed

Example GUI Screenshot of GUI version of the application. Notice the MAXINCIDENCE parameter is changed from the default value of 90.0.

Input image

Example input cube Parameter Name: FROM Screenshot of the input image before photometric correction has been performed.

Example input PVL	Pvl file including the input photometric parameters for the input cube.

Output image

Example output cube Parameter Name: TO Screenshot of the output image after photometric correction. Notice there are different areas of valid pixels from Examples 1 and 2 output images. This happens since the pixel values have been set to null where the incidence angles are greater than 85, but the emission angle parameter is kept at default.

This example shows the photometric correction of a cube by using photomet application with the maximum emission angle set to 75.0 and the maximum incidence angle set to 85.0.

photomet photomet from=input.cub phopar=input.pvl to=output.cub maxemission=75.0 maxincidence=85.0

photomet GUI with MAXEMISSION and MAXINCIDENCE changed

Example GUI Screenshot of GUI version of the application. Notice the MAXEMISSION and MAXINCIDENCE parameters are changed from the default values of 90.0.

Input image

Example input cube Parameter Name: FROM Screenshot of the input image before photometric correction has been performed.

Example input PVL	Pvl file including the input photometric parameters for the input cube.

Output image

Example output cube Parameter Name: TO Screenshot of the output image after photometric correction. Notice this output image contains the overlapping valid pixels from Examples 2 and 3.