 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> It is my code.Real image is a photograph.the value of each pixel in real image
> are like these:
> 0.220900
> 0.375800
> 0.375800
> 0.346500
> 0.277700
> 0.355800
> 0.396900
> 0.441800
> 0.442800
> 0.382300
> I doublt this statement "pigment {color rgb spectral}"is wrong.spectral is
> responsible to "pov-ref.txt" which save the value of each pixel of real image.
In general, writing "pigment{color rgb X}" does not mean the output
image will have a pixel value of X. The value X tells POV how diffusely
reflective the surface is, POV then uses the brightness of any lights,
the relative position of the objects and surface normals to calculate
what the pixel colour should be.
If your goal is to match the brightness of POV's output with your photo,
then I suggest you simply add a multiplier to your light source and
tweak it until they match:
#local FACTOR = 10.0;
light_source {<sun_x,sun_y,sun_z>
color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>*FACTOR }
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
CAS <sup### [at] cstnet cn> wrote:
> I have known the value of solar irradiance in 817nm band is 1.08279
> W/(m2*nm).How can I define it? "light_source {<sun_x,sun_y,sun_z> color rgb
> 1.08279 } "I set statement as above in pov,but I doublt it is wrong,because
> 1.08279 is a solar irradiance,color rgb is color ，but I donnot know how
> to convert them or how to define solar irradiance.beg a hot to help me !thank
> you!
Note that for the most part POV-Ray uses a quite simplistic Phong
lighting model with RGB values. It does not model light at different
wavelengths, nor is it able to process colors in any more complicated
ways than manipulating RGB values, nor are most of its lighting
calculations any more complex than the simple Phong lighting model.
(Note that "Phong lighting model", or "Phong reflection model", should
not be confused with "phong highlights". The lighting model is an
illumination model that simply adds three components to the color of
a surface: ambient (a constant), diffuse (a lighting component that's
independent of the viewing angle) and specular (a lighting component
that's dependent on the viewing angle).)
Far from me to disparage POV-Ray, but to be completely honest, the
Phong lighting model is quite poor and simplistic. (It's *major*
advantage is that it's very fast and can produce, when properly used,
acceptable results.)
If you want to convert a unit of irradiance into RGB, you'll have to
figure out the conversion formula. This is probably not a trivial task,
nor is it probably very productive because of the limitations of the
lighting model described above.
(There are many problems in getting an "accurate" conversion from some
standard irradiance unit to RGB. One of them is that the relation between
irradiance and the apparent brightness of what ends up on your monitor
screen is subject to rather complex gamma curve functions, which are not
only dependent on the gamma used by POV-Ray and the image file, but also
the gamma correction performed by your display hardware and the screen.)
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp <war### [at] tagpovrayorg> wrote:
> CAS <sup### [at] cstnetcn> wrote:
> > I have known the value of solar irradiance in 817nm band is 1.08279
> > W/(m2*nm).How can I define it? "light_source {<sun_x,sun_y,sun_z> color rgb
> > 1.08279 } "I set statement as above in pov,but I doublt it is wrong,because
> > 1.08279 is a solar irradiance,color rgb is color ，but I donnot know how
> > to convert them or how to define solar irradiance.beg a hot to help me !thank
> > you!
>
> Note that for the most part POV-Ray uses a quite simplistic Phong
> lighting model with RGB values. It does not model light at different
> wavelengths, nor is it able to process colors in any more complicated
> ways than manipulating RGB values, nor are most of its lighting
> calculations any more complex than the simple Phong lighting model.
>
> (Note that "Phong lighting model", or "Phong reflection model", should
> not be confused with "phong highlights". The lighting model is an
> illumination model that simply adds three components to the color of
> a surface: ambient (a constant), diffuse (a lighting component that's
> independent of the viewing angle) and specular (a lighting component
> that's dependent on the viewing angle).)
>
> Far from me to disparage POV-Ray, but to be completely honest, the
> Phong lighting model is quite poor and simplistic. (It's *major*
> advantage is that it's very fast and can produce, when properly used,
> acceptable results.)
>
> If you want to convert a unit of irradiance into RGB, you'll have to
> figure out the conversion formula. This is probably not a trivial task,
> nor is it probably very productive because of the limitations of the
> lighting model described above.
>
> (There are many problems in getting an "accurate" conversion from some
> standard irradiance unit to RGB. One of them is that the relation between
> irradiance and the apparent brightness of what ends up on your monitor
> screen is subject to rather complex gamma curve functions, which are not
> only dependent on the gamma used by POV-Ray and the image file, but also
> the gamma correction performed by your display hardware and the screen.)
>
> --
> - Warp
I have tryed to modify my code "light_source {<sun_x,sun_y,sun_z>
color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>*10}"or *100,I
find the result is the same as the "light_source {<sun_x,sun_y,sun_z>
color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>".
what I know is the value of reflection of each sphere .this reflection is amount
to specular (a lighting component that's dependent on the viewing
angle)."pigment {color rgb spectral}"maybe this sentence is wrong,this sentence
I defined in order to describe the spectral of each sphere.Maybe I should use
"finish {spectral X}".
I know pov is a render ,if I define light_source ,camera,objects,it will get a
image,I am not quite clear how pov to calculate color of each pixel,only use
photon model?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
scott <sco### [at] scottcom> wrote:
> > It is my code.Real image is a photograph.the value of each pixel in real image
> > are like these:
> > 0.220900
> > 0.375800
> > 0.375800
> > 0.346500
> > 0.277700
> > 0.355800
> > 0.396900
> > 0.441800
> > 0.442800
> > 0.382300
> > I doublt this statement "pigment {color rgb spectral}"is wrong.spectral is
> > responsible to "pov-ref.txt" which save the value of each pixel of real image.
>
> In general, writing "pigment{color rgb X}" does not mean the output
> image will have a pixel value of X. The value X tells POV how diffusely
> reflective the surface is, POV then uses the brightness of any lights,
> the relative position of the objects and surface normals to calculate
> what the pixel colour should be.
>
> If your goal is to match the brightness of POV's output with your photo,
> then I suggest you simply add a multiplier to your light source and
> tweak it until they match:
>
> #local FACTOR = 10.0;
> light_source {<sun_x,sun_y,sun_z>
> color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>*FACTOR }
I have modified my code:
" #local FACTOR = 10.0;
> light_source {<sun_x,sun_y,sun_z>
> color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>*FACTOR }"
I find the result is the same as "light_source {<sun_x,sun_y,sun_z>
> color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>",nor I mutiply 10
or 100,the result is not change.I
am so confuse!
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"CAS" <sup### [at] cstnetcn> wrote:
> scott <sco### [at] scottcom> wrote:
> > > It is my code.Real image is a photograph.the value of each pixel in real image
> > > are like these:
> > > 0.220900
> > > 0.375800
> > > 0.375800
> > > 0.346500
> > > 0.277700
> > > 0.355800
> > > 0.396900
> > > 0.441800
> > > 0.442800
> > > 0.382300
> > > I doublt this statement "pigment {color rgb spectral}"is wrong.spectral is
> > > responsible to "pov-ref.txt" which save the value of each pixel of real image.
> >
> > In general, writing "pigment{color rgb X}" does not mean the output
> > image will have a pixel value of X. The value X tells POV how diffusely
> > reflective the surface is, POV then uses the brightness of any lights,
> > the relative position of the objects and surface normals to calculate
> > what the pixel colour should be.
> >
> > If your goal is to match the brightness of POV's output with your photo,
> > then I suggest you simply add a multiplier to your light source and
> > tweak it until they match:
> >
> > #local FACTOR = 10.0;
> > light_source {<sun_x,sun_y,sun_z>
> > color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>*FACTOR }
>
> I have modified my code:
> " #local FACTOR = 10.0;
> > light_source {<sun_x,sun_y,sun_z>
> > color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>*FACTOR }"
> I find the result is the same as "light_source {<sun_x,sun_y,sun_z>
> > color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>",nor I mutiply 10
or 100,the result is not change.
I
> am so confuse!
"#version 3.7;
//camera information:
camera
{
angle 7.54371
location <camera_x,camera_y,camera_z >
look_at <0,0,0>
rotate <0,-31,0>
right x
up y
}
//light information:
light_source {<sun_x,sun_y,sun_z>
color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>
}
//scene information:
#declare geom_file_name = "pov-xyz.txt";
#declare spec_file_name = "pov-ref.txt";
#declare n=o;
#fopen Input_geom_file geom_file_name read
#fopen Input_spec_file spec_file_name read
#while(defined(Input_geom_file Input_spec_file))
#read(Input_geom_file,x1,y1,z1)
#read(Input_spec_file,ref)
sphere
{
<x1-center_x,y1-center_y,z1-center_z>, //sphere location
1.55 //sphere radius
finish {specular ref} //sphere specular
}
#declare n = n + 1;
#end
"
Above is all my code in pov,my aim is to use illumination model (phong light
model) to calculate the light intensity of each point.I want to know if I want
to use phong model to calculate ,whether need I to define phong to tell pov to
calculate.my code is all above,how can I to define phong model to calculate,I
need to use specular to calculate light intensity of each pixel. And "finish
{specular ref}" Is my write right? Beg a help!
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> I have modified my code:
> " #local FACTOR = 10.0;
>> light_source {<sun_x,sun_y,sun_z>
>> color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>*FACTOR }"
> I find the result is the same as "light_source {<sun_x,sun_y,sun_z>
>> color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>",nor I mutiply 10
or 100,the result is not change.I
> am so confuse!
You will probably need to make FACTOR very large to see a difference,
like 1000000 as your rgb components are very small.
Also POV adds ambient light by default to every shape, which is I think
what you are seeing in the output. You can turn it off by adding this
line at the top of your code:
#default{ finish{ ambient 0 }}
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp <war### [at] tagpovrayorg> wrote:
> CAS <sup### [at] cstnetcn> wrote:
> > I have known the value of solar irradiance in 817nm band is 1.08279
> > W/(m2*nm).How can I define it? "light_source {<sun_x,sun_y,sun_z> color rgb
> > 1.08279 } "I set statement as above in pov,but I doublt it is wrong,because
> > 1.08279 is a solar irradiance,color rgb is color ，but I donnot know how
> > to convert them or how to define solar irradiance.beg a hot to help me !thank
> > you!
>
> Note that for the most part POV-Ray uses a quite simplistic Phong
> lighting model with RGB values. It does not model light at different
> wavelengths, nor is it able to process colors in any more complicated
> ways than manipulating RGB values, nor are most of its lighting
> calculations any more complex than the simple Phong lighting model.
>
> (Note that "Phong lighting model", or "Phong reflection model", should
> not be confused with "phong highlights". The lighting model is an
> illumination model that simply adds three components to the color of
> a surface: ambient (a constant), diffuse (a lighting component that's
> independent of the viewing angle) and specular (a lighting component
> that's dependent on the viewing angle).)
>
> Far from me to disparage POV-Ray, but to be completely honest, the
> Phong lighting model is quite poor and simplistic. (It's *major*
> advantage is that it's very fast and can produce, when properly used,
> acceptable results.)
>
> If you want to convert a unit of irradiance into RGB, you'll have to
> figure out the conversion formula. This is probably not a trivial task,
> nor is it probably very productive because of the limitations of the
> lighting model described above.
>
> (There are many problems in getting an "accurate" conversion from some
> standard irradiance unit to RGB. One of them is that the relation between
> irradiance and the apparent brightness of what ends up on your monitor
> screen is subject to rather complex gamma curve functions, which are not
> only dependent on the gamma used by POV-Ray and the image file, but also
> the gamma correction performed by your display hardware and the screen.)
>
> --
> - Warp
Thanks for your help.I am so sorry to trouble you again. I still have some
confusions:
1:how does pov calculate the value of each pixel of final image? My
understanding is :using the irradiance and surface properties to calculate the
value of each pixel of final image. Using the simplistic Phong lighting model to
calculate the color of each pixel of final image.I didnot set related "phong"
statements but also get a result,my codes are given below,I beg you have a check
whether it is true.Emphasisly,my reflection is BRDF（dependent on suning
angle、 viewing angle 、wavelength）.I use "specular X".Is it
right?
2:I know irradiance 1.08279（W/(m2*nm).I have converted XYZ to the values
of color rgb using formula,my conversion formula is:
r=(3.2406*3.815429E-06)+(-1.5372*1.523114E-06)
g=(-0.9689*3.815429E-06)+(1.8758*3.815429E-06)
b=(0.0557*3.815429E-06)+(-0.2040*3.815429E-06)
but the value of b is negative,it must be something wrong in it.I only know the
solar irradiance is 1.08279（W/(m2*nm),not know what color the
light_source,so how can I code with it? "color rgb 1.08279 "must be wrong?
3:HDR image is a image of scene,its dynamic range is generally more than 100: l,
and record each pixel which value is real scene brightness (CD/m^2).so I regard
it as radiance.is it right?
miaoyu
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> 2:I know irradiance 1.08279（W/(m2*nm).I have converted XYZ to the values
> of color rgb using formula,my conversion formula is:
> r=(3.2406*3.815429E-06)+(-1.5372*1.523114E-06)
> g=(-0.9689*3.815429E-06)+(1.8758*3.815429E-06)
> b=(0.0557*3.815429E-06)+(-0.2040*3.815429E-06)
> but the value of b is negative,it must be something wrong in it.
As I said in another post, if b is negative this indicates that the
colour of your light source is outside the sRGB colour space. There is
nothing physically wrong with that, sRGB only represents a portion of
the colours we can see. Your light source is 817nm which is well into
infra-red, a camera will see it but humans will struggle - that's why
your calculated rgb values are so tiny and negative.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> scott <sco### [at] scottcom> wrote:
>>> It is my code.Real image is a photograph.the value of each pixel in real image
>>> are like these:
>>> 0.220900
>>> 0.375800
>>> 0.375800
>>> 0.346500
>>> 0.277700
>>> 0.355800
>>> 0.396900
>>> 0.441800
>>> 0.442800
>>> 0.382300
>>> I doublt this statement "pigment {color rgb spectral}"is wrong.spectral is
>>> responsible to "pov-ref.txt" which save the value of each pixel of real image.
>>
>> In general, writing "pigment{color rgb X}" does not mean the output
>> image will have a pixel value of X. The value X tells POV how diffusely
>> reflective the surface is, POV then uses the brightness of any lights,
>> the relative position of the objects and surface normals to calculate
>> what the pixel colour should be.
>>
>> If your goal is to match the brightness of POV's output with your photo,
>> then I suggest you simply add a multiplier to your light source and
>> tweak it until they match:
>>
>> #local FACTOR = 10.0;
>> light_source {<sun_x,sun_y,sun_z>
>> color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>*FACTOR }
>
> I have modified my code:
> " #local FACTOR = 10.0;
>> light_source {<sun_x,sun_y,sun_z>
>> color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>*FACTOR }"
> I find the result is the same as "light_source {<sun_x,sun_y,sun_z>
>> color rgb <1.00229e-005,3.46021e-006,-5.65828e-007>",nor I mutiply 10
or 100,the result is not change.I
> am so confuse!
>
>
By default, every textures have this:
finish{diffuse 0.7 ambient 0.1}
Your red component, the largest one, is 10000 times smaller than the
ambient value.
With such a disproportion, it's normal that you don't see any
difference. By multiplying your light intensity by 10 to 100, you bring
the light's intensity from 1/10000 to 1/1000 up to 1/100 of the ambient
level.
Alain
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
>> 2:I know irradiance 1.08279（W/(m2*nm).I have converted XYZ to
>> the values
>> of color rgb using formula,my conversion formula is:
>> r=(3.2406*3.815429E-06)+(-1.5372*1.523114E-06)
>> g=(-0.9689*3.815429E-06)+(1.8758*3.815429E-06)
>> b=(0.0557*3.815429E-06)+(-0.2040*3.815429E-06)
>> but the value of b is negative,it must be something wrong in it.
>
> As I said in another post, if b is negative this indicates that the
> colour of your light source is outside the sRGB colour space. There is
> nothing physically wrong with that, sRGB only represents a portion of
> the colours we can see. Your light source is 817nm which is well into
> infra-red, a camera will see it but humans will struggle - that's why
> your calculated rgb values are so tiny and negative.
>
In this case, it may advisable to work on so called "false colours"
where the actual spectrum is shifted and stretched or compressed to make
it fit within the visible range.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
