►
From YouTube: SJAA Imaging SIG Meeting August 18th 2020 Multi-Level Processing: A Live Demonstration
Description
This month we're very excited to have Eric Coles presenting on:
The Hubble Palette – Mapped Colors of Emission Nebula with Multi-Level Processing: A Live Demonstration
Eric is a phenomenal astrophotographer and a master of image processing.
Please check out his gallery at https://www.astrobin.com/users/coles44/
A
I
want
to
welcome
everybody
to
the
august
2020
san
jose
astronomical
association,
imaging
special
interest
group
meeting.
We
are
very
happy
and
feel
very
privileged
to
have
eric
cole's
presenting
tonight
and
eric
has
some
phenomenal
new
ways
of
of
processing
using
photoshop,
and
I
think
with
that
I'll
just
turn
it
over
to
you.
Eric.
B
Okay,
thanks
bruce
well,
I
know
if
they're
phenomenal,
but
they've
been
interesting
and
what
I've
tried
to
do,
I'm
primarily
an
imager
for
emission
nebula
and
another
reflection
nebula
and
in
the
hubble
palette.
Images
is
a
standard
mapping
and
show
mapping,
and
it
was
a
standard
procedure
mostly
copied
from
j.p
mitzvanio
who's,
really
an
expert
in
this
area,
but
in
the
last
few
months
I've
kind
of
developed
another
technique
which
brings
out
more
of
the
color
variation
in
different
parts
of
the
nebula.
B
B
B
B
It
just
goes
from
one
side
to
the
other
and
is
again
a
real
pleasure
to
work
with
it.
This
whole
setup
really
disappears,
so
I
start
it
in
the
evening
and
come
back
and
collect
data
in
the
morning.
You
can
see
it's
mounted
on
a
wedge
and
this
big
boat
anchor
you
see
in
the
bottom
was
an
offset
pier.
We
need
to
have
the
offset
pierce,
otherwise
it
would
have
hit
the
wall
and
you
can
see
it's
pointed
against
a
flat
screen.
So
I
take
my
flats
when
I
change
rotation
in
the
camera
and
software.
B
B
B
B
B
B
B
B
Normally
we
take
the
stacks
and
we'll
go
into
pix
insight,
we'll
take
the
linear
files
we'll
create
the
non-linear
files
will
take
care
of
all
the
noise
and
then
we'll
create
a
second
file
which
has
no
stars,
because
what
we're
going
to
do
here
with
this
show
map
processing
is
do
tone.
Mapping
and
the
tone
mapping
is
done
with
the
starless
images
and
then
a
luminosity
layer
is
put
in
on
top
of
that.
B
So
the
first
image
I'm
going
to
process
is
called
the
tarantula
nebula
and
I'm
going
to
open
up
four
files,
the
hydrogen
alpha,
which
has
stars
in
it
the
hydrogen
alpha
that
doesn't
have
stars
in
it,
the
starless
image
and
the
starless
image
for
the
oxygen
and
sulfur.
In
this
case,
I
don't
really
need
the
hot,
the
sulfur
and
oxygen
images
with
stars,
because
we
are
doing
tone
mapping.
B
B
So
this
is
a
tarantula,
the
hydrogen
alpha
with
stars,
and
then
we
have
them
in
order
that
we're
going
to
map
them
to
rgb
the
sulfur,
the
oxygen
and
well,
because
a
little
bit
out
it'll
be
sulfur,
move
this
over
hydrogen
alphas
and
oxygen.
You
might
notice
that
you
probably
have
you
say
you
haven't
removed
all
the
stars
here,
but
those
are
actually
small
star
clusters.
B
So
you
see
we
start
off
where
all
the
channels
are
really
just
the
the
sulfur
starless
image.
So
now
I'm
going
to
get
data
from
the
hydrogen
alpha
again
mark
all
copy,
put
it
into
our
new
image
and
put
that
in
the
green.
So
now
we're
mapping
hydrogen
alpha
to
the
green
channel,
paste
it
in
go
back
and
get
the
oxygen
image
copy,
move
it
over
into
our
new
image
and
I'll
put
that
in
the
blue
channel.
B
So
the
standard
shl
mapping
is
sulfur
is
red.
Hydrogen
alpha
is
green
and
oxygen.
Three
is
blue
and
here's
our
starting
image.
Let's
go
to
the
layers.
Now
you
might
say
that,
that's
you
know,
that's
not
really
a
bad
image.
Maybe
we
should
just
stop
there,
but
there
are
different
levels
of
intensity
in
this
image
and
the
idea
in
this
kind
of
tone,
mapping
sho
image,
is
to
balance
all
the
three
contributions
approximately
equally
in
most
images,
perhaps
not
in
this
one,
the
hydrogen
alpha
will
really
dominate
and
this
image
will
be
green.
B
So
we
don't
want
that.
So
normally,
what
someone
will
do
is
simply
go
in.
Let
me
bring
up
the
histogram
and
just
balance
these
three
and
then
you're
done
with
it.
You
put
on
your
luminosity
layer
and
that's
the
end
of
your
image,
but
I
really
figured
out
that
you
can
get
a
better
image
and
more
variation
in
color.
B
And
you
might
say
that
your
selections
are
pretty
ragged
and
actually
that's
the
technique.
You
don't
want
them
to
be
too
smooth,
because
if
they
are,
then
you
will
see
the
interface
between
the
areas
that
you've
selected
and
to
do
this
even
further.
I'm
going
to
feather
this,
but
first
I'm
going
to
select
the
inverse
so
now
I've
selected
this
dark
area
here
and
now
I'm
going
to
feather
this.
B
Make
it
even
fuzzier
250
pixels
and
now
we're
going
to
get
rid
of
those
darn
marching
ants.
I
always
hate
the
marching
ants.
So
let's
just
hide
the
marching
ants,
but
we
still
have
selected
this
area
versus
this
unselected
area
and
you
can
see
the
histogram
really
isn't
balanced,
so
I'm
going
to
go
in
and
balance
them
kind
of
visually
the
best
I
can
by
getting
adjustments
and
levels
I'll
move
this
over.
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
You
can
interrupt
me,
so
we
we
don't
have
any
stars.
This
is
just
our
tone
map
the
one
thing
to
mention.
If
we
tried
to
make
these
adjustments
with
the
starred
images,
as
we
spread
out
to
histogram,
your
stars
would
get
larger
and
larger
and
larger
or
smaller
and
you'd
end
up
with
stars
with
different
colored
rings
around
it,
and
when
you
just
put
your
show
images
together
and
try
to
make
adjustments,
you
end
up
with
really
funky
looking
funky
looking
stars
and
in
any
case
you
know
the
star.
B
B
So
let's
take
the
start
image
I'll
make
a
copy
of
that,
and
this
is
going
to
become
our
luminosity
layer,
we'll
put
it
on
top,
and
this
is
called
hydrogen
luminosity,
and
we
will
merge
this
with
the
layer
below
as
a
luminosity
layer.
Now,
when
you
do
this,
it
always
ends
up
being
a
little
bit
too
bright.
So
what
I'm
going
to
do
is
just
adjust
the
levels.
B
And
there's
our
image
without
stars
and
wood
stars,
but
there's
more
detail
to
get
out
of
this
than
than
just
using
this
as
luminosity
and
there's
a
technique
which
we
use
called
high,
pass
filtering
and
what
this
does.
If
you
know
anything
about
electronics,
it
used
to
be
that
you
put
in
high-pass
filters
to
get
rid
of
all
the
noise.
B
So
you
could
just
select
the
signal
that
you're
interested
in
well
high
pass
filtering
when
you're
doing
this
kind
of
imaging
is
not
really
the
same,
but
it
cuts
the
histogram
in
a
way
which
allows
you
to
show
the
edges
of
each
of
these
structures
by
the
way.
There's
one
more
thing
that
I'd
like
to
do
and
just
see
how
this
works
out,
I'm
going
to
make
another
copy.
There's
one
adjustment.
It's
called
the
bob
frankie
adjustment,
so
we're
going
to
label
this
as
bf.
B
B
B
B
B
Turn
them
all
off,
except
one
and
I'm
going
to
zoom
in
on
this
kind
of
detailed
area
right
here
and
see
whether
I
could
highlight
the
edges
of
some
of
these
structures
to
do
so
as
to
bring
out
more
detail
and
the
way
we
do
this
is
we
select
filters,
others
high
pass
and
we're
going
to
make
a
two
pixel
pass
and
let's
look
at
the
histogram
and
you
see
what
we've
done.
We've
actually
cut
it
more
than
you
see
right
here.
B
But
if
you
look
closely,
you
can
see
these
little
lines,
I'm
not
sure
if
it
comes
across
in
the
video.
But
let
me
zoom
in
I
can't
zoom
in
right
now,
so
we're
going
to
do
a
high
pass
filter,
and
now
you
see,
we've
cut
everything
except
2,
pixel
wide
right
in
the
center
of
the
histogram,
and
now
I'm
going
to
overlay
it
on
the
top
of
the
other
image.
B
B
B
Our
tone
map
our
tone
map
with
luminosity
and
our
tone
map
with
a
high
pass
filter.
So
there
is
our
final
hubble
palette
image
with
stars
now.
The
interesting
thing
is,
you
don't
have
to
have
stars
in
there
and
you
can
still
have
kind
of
an
abstract
image
which
some
people,
including
myself,
find
interesting.
So
if
I
turn
off
the
luminosity
layer,
I
have
a
little
abstract
piece
of
art
in
the
hubble
palette
and
oftentimes.
I
will
post
that,
along
with
the
one
with
stars.
A
B
Yeah,
it's
a
green
to
gold
if
you
know,
but
you
probably
know
who
bob
frankie
is:
we've
exchanged
data
now
for
years
and
he
has
a
little
adjustment
and
I
kind
of
like
it,
and
so
at
the
end
of
all
these
adjustments,
I
usually
put
in
a
few
percentage
of
the
bob
frankie
adjustment.
It's
just
as
simple
going
in
and
changing
the
colors
a
little
bit
and
a
good
hubble
palette
image
should
go
from
the
greens
all
the
way
to
the
golds
and
the
purple.
B
C
B
C
B
B
B
So
sometimes
I
just
turn
those
down,
so
you
have
more
of
a
hint
of
stars
without
them
really
dominating
the
image
and
it
doesn't
dominate
the
image
in
this
particular
nebula,
because
the
nebula
is
so
darn
bright,
but
in
some
sometimes
it
will,
and
so
you
want
to
kind
of
just
tone
my
tone
down
your
stars
and
there's
right,
there's
different
ways
of
doing
it.
You
can
do
it
with
opacity.
B
D
A
D
B
Yeah
they're,
they
are
not
visible.
So
the
only
point
of
doing
all
these
different
layers
was
to
illustrate
that,
as
I
select
different
parts
of
the
image
and
adjust
the
histogram
you,
you
get
a
different
result
as
opposed
to
adjusting
the
histogram
of
the
entire
image
in
the
beginning,
which
is
what
most
people
do.
D
B
B
D
B
C
B
I'll
do
some
noise
reduction.
In
fact,
on
this
image
I
might
go
in
and
do
some
noise
reduction,
but
you
know
that's
that's
kind
of
a
waste
of
our
time.
I
don't
think
we
really
have
to
do
that,
but
there
are
some
kind
of
noisy
areas
down
here
which
I
could
clean
up
any
more
questions,
or
should
we
go
on
to
the
next
one.
B
B
And
as
far
as
color
and
beauty,
this,
this
nebula
in
itself
is
kind
of
a
work
of
art.
If
you
ever
get
your
hands
on
this,
this
data,
you
really
cannot
go
wrong
any
way.
You
process
it.
It's
going
to
come
out
to
be
really
beautiful,
and
so
let's
just
do
the
same
thing.
I
don't
think
I
have
to
detail
everything
as
before,
but
let's
start
out
with
our
hydrogen
alpha
starless
image,
I'm
going
to
copy
it.
B
And
sulfur,
you
probably
realize
that
I
did
it
some
processing
before
we
got
to
this
point
and
the
processing,
besides
removing
the
stars,
was
really
to
stretch
out
the
sulfur
and
the
oxygen.
So
I
don't
have
to
do
so
much
processing
once
I
get
in
to
do
the
hubble
pellet.
But
if,
if
I
took
these
the
way
they
normally
were
they'd
be
a
lot
darker,
but
getting
rid
of
the
stars
allows
me
to
stretch
that
histogram
so
that
it
is
more
or
less
balanced,
even
from
the
start.
B
B
And
these
are
balanced
by
eye
there's
no
anal
analytics
involved.
I
just
look
at
the
histogram
and
try
to
get
as
much
overlapping
area
as
possible.
Now
I
can
tell
you
that
people
that
work
with
this
nebula
look
at
this
image
right
here
and
say:
well,
that's
good!
Why
should
we
go
any
further?
But
if
you
do
a
multi-level
adjustment,
you
will
find
out.
You
will
bring
more
variation
of
color
and
really
that's
the
whole
objective
of
of
doing
a
hubble
palette.
B
B
B
B
B
B
B
B
B
B
B
B
B
B
I
think
that
looks
better,
so
you
can
see
once
again
that
by
going
in
and
selecting
specific
areas,
adjusting
the
histogram
in
a
standard,
hubble
palette
way
where
you
try
to
balance
the
three
emissions
you
end
up
with
a
more
interesting
image.
So,
let's
turn
on
our
stars
and
next
thing:
we're
going
to
do
is
get
a
little
more
detail.
B
B
A
B
It
has
to
do
with
the
quality
of
the
image,
so
everything
is
different.
If
you,
if
you
really
don't
this,
has
a
lot
of
resolution,
so
I
can
start
out
with
really
small
pixels.
If
you
don't
have
great
resolution
in
your
starless
image,
you'll
start
out
a
higher
number.
Everything
is
kind
of
done
by
feel
there
is
no
right
or
wrong
and
the
larger
number
you
you
choose
for
a
high
pass
filter,
the
more
the
larger
the
structure
you're
going
to
emphasize.
B
B
B
B
Let's
let
that
save
and
there's
our
hubble
palette
image
of
the
edicarina
nebula,
with
a
little
more
emphasis
on
the
the
keyhole
and
again
you
can
turn
off
luminosity
and
this
becomes
kind
of
an
abstract
work
of
art,
and
it's
actually,
let's
see
do
I
have
it
up
here.
No,
I
didn't
bring
it
up
it's
in
the
basement.
B
B
B
You
can
map
them
just
as
regular
images
select
those
stars
by
making
a
starless
image
and
put
it
on
top
or
you
can
just
do
an
rgb
image
and
put
those
in
here.
But
when
you
do
it,
what
you
find
out
is
that
rgb
stars
are
big
compared
to
narrowband
stars.
These
narrowband
stars
are
tiny
little
things,
they're
little
dots,
no
matter
how
good
your
data
for
an
rgb
star,
it's
going
to
be
big
and
kind
of
blotchy
and
you're
not
going
to
like
it,
and
the
second
thing
is
rgb
stars.
B
The
color
scheme
is
completely
different
than
a
hubble
palette
and
it
kind
of
it's
hard
to
look
at.
It
doesn't
look
very
good.
So
the
answer
is
I've
done
it.
I've
experimented
with
it,
but
I
always
kind
of
say
blah
and
turn
off,
but
yeah.
It's
fairly
easy
to
do.
You
can
do
it
with
narrow
band
data
to
get
artificial,
colored
stars
or
you
can
do
it
with
rgb
data
and
then
just
bring
the
stars
in
it's
a
fairly
simple
process,
but
star
color
really
has
no
meaning
in
narrow
band
image.
B
A
E
B
B
B
B
B
C
B
B
B
B
B
B
And
now
you've
lit
up,
so
you
can
see
all
the
the
oxygen
emissions
and
that's
how
I
add
them
into
an
rgb
image.
So
this
is
without
and
then
with,
and
you
can
change
how
much
you
apply
that
you
can
you
don't
have
to
apply
it
a
hundred
percent,
but
in
any
rgb,
where
I
take
the
hydrogen
alpha,
the
emission
nebula.
B
I
always
generate
a
starless
image
if
there's
enough
oxygen
I'll
do
a
blue
one
and
if
there's
of
course
there's
enough
hydrogen,
I
will
do
a
red
one
and
light
them
up
and
basically
it's
just
like
shining
a
light
on
the
emission
nebula,
and
without
that
you
can
still
see
you
can
still
see
where
the
emissions
are.
But
it's
not
as
dramatic.
B
Yeah,
the
problem
is,
if
you
try
to
add
in
just
I
know,
other
people
will
take
the
the
hydrogen
and
mix
it
in
with
the
red
that
you've
gathered
as
in
your
our
filter,
but
it's
it's
hard
to
control
the
look
of
the
stars
that
way.
So,
when
you're
working
with
starless
image,
it's
very
easy
to
control
and
if
you
think,
there's
too
much,
you
know
too
much
lightning
there.
You
can
just
simply
change
this
to
well,
that's
a
little
too
much
red
so
I'll
make
it
50
I'll
get
rid
of
that.
B
B
B
B
F
B
So
if
you're
someone
like
warren,
he
does
all
his
processing
in
pi,
and
god
love
him.
For
that.
I
can't
do
that.
So
all
the
pre-processing,
starting
from
my
subs,
all
the
pre-processing,
is
done
in
pi.
All
the
initial
stretching
and
adjustments
of
color
are
done
in
pi.
All
the
noise
reduction
is
done
in
pi
and
what
else?
B
And
why
do
I
do
it
that
way?
Well,
for
one
thing:
I've
been
working
with
photoshop
for
30
years,
so
I'm
not
willing
to
flush
that
all
away
and
pi
is
it's
really
good
in
some
of
the
you
know
the
grand
processes-
and
I
think
it's
unmatched
in
things
like
all
the
pre-processing
and
there's
no
better
program
to
do
that.
B
B
I
end
up
using
not
just
tg
video
noise,
which
can
get
a
little
blotchy
on
some
really
weak
images
or
a
stretch,
but
I
use
mmt
with
with,
I
think,
seven
or
eight
different
layers
just
to
smooth
it
out,
because
once
I
get
into
photoshop
I
I
really
don't
care.
You
know
the
detail
is
going
to
come
from
the
hydrogen
alpha.
B
I'm
just
trying
to
get
the
color
in
from
those
weaker
images,
so,
like
80
percent,
is
done
with
tgv,
denoise
and
then
sometimes
really
noisy,
blotchy
images,
I'll
use
mmt,
and
there
was
a.
I
think,
the
astro
imaging
channel
has
really
a
nice.
We
did
a
nice
session
on
noise
reduction
which
which
describes
how
best
to
use
both
of
those
processes.
A
A
C
B
B
B
B
B
So
now,
if
I
copy
the
oxygen
it's
already
marked
at
all
copy,
I
go
to
the
new
document.
I
select
the
blue
channel
and
I
paste
it
in
now.
The
blue
channel
is
oxygen
and
I
select
the
sulfur
image
copy.
It
command
c.
Go
to
my
new
document.
Select
the
red
channel
paste
it
in
so
now
I
put
the
sulfur
in
the
red
channel,
the
hydrogen
l
from
the
green
channel.
B
C
B
C
B
This
is
kind
of
the
way.
This
is
a
variation
on
the
way
jp.
Does
it
basically
puts
him
into
different
channels
and
then
starts
working
with
histograms?
He
actually
does
some
other
things
now,
which
you
know
I
I
may
or
may
not
agree
with,
and
he
doesn't
do
this
kind
of
multi-layer
adjustment.
But
if
you
know
jp
metsylvania
I'll
never
see
his
images,
they
are
stupendous,
they
are
just
works
of
art,
all
of
them.
B
He's
astro
anarchy.
I
I've
tried
that,
but
it's
so
much
work
and
you
end
up
with
10
seconds
worth
of
video.
It
just
really
didn't
appeal
to
me.
It's
I
will.
I
will
leave
that
to
him.
It's
it's
a
tremendous
amount
of
work
to
put
those
different
layers
and
then
to
fly
through
besides
the
people
that
do
cgi.
Do
that
much
better!
B
The
other
thing
that
he
might
be
talking
about,
if
I
might
be
doing
a
close-up
of
say
the
elephant
trunk,
and
I
will
just
feather
that
in
in
photoshop
and
that's
just
process,
so
I
won't
try
to
bring
that
in
in
in
the
stack
say
of
hydrogen
alpha
but
I'll.
Take
the
hydrogen
alpha
stack
process
that
separately
with
a
separate
field
of
view,
register
it
to
the
the
entire
field
of
view
and
then
feather
it
in
and
try
to
to
match
the
the
brightness
in
photoshop.
B
B
B
B
B
E
H
Just
a
speck,
I'm
just
a
snack
compared
with
a
star.
The
planet
is
just
another
spectrum
to
think
about
all
of
this
to
think
about
the
best
emptiness
of
space,
millions
and
millions
of
stars,
billions
and
billions
of
specs.
The
beauty
of
the
living
thing
is
not
the
atoms
that
go
into
it,
but
the
way
those
atoms
are
put
together.
I
F
I
I
B
B
I
thought
maybe
I
was
running
now:
it's
not
going
to
start
running
for
another
another
10
minutes
we're
imaging
tonight.
If
all
goes
well,
we're
going
to
complete
sharpless112
I've
already
done
the
hubble
palette.
Now
we're
going
to
do
the
the
rgb
and
put
it
and
lighten
them
up
with
the
hydrogen
alpha
and
the
oxygen
which
are
the
two
primary
emissions
from
that
nebula.
From
that
sharpless
object.