Can I check something with you guys

On Tue, 27 Nov 2007 22:22:59 -0500, Andrew Kalten
wrote:

Glenn wrote:
Is this glaringly obvious to you guys as well.


The only way to answer this question objectively is
to subtract one image from the other.

Using Linux/GNU software (it would be very hard to accomplish
on a stock MS Windows system) I converted both JPEG images
to an uncompressed format, subtracted the image data byte
by byte, and then recompressed back to JPEG (to get a smaller
file size for posting).

The subtraction will reveal any discrepancy. If the images
are identical, the values will subtract to zero and the appearance
will be black. The result is attached.

As can be seen, the difference image is virtually all black, except
for a slightly visible band of lightness corresponding to the water
areas. This slight difference, only barely perceptible, is likely the
result of different compression qualities between the two images.

Conclusion: The images are virtually identical. Any difference will
be beyond perception.

AK


Hello,
In addition, IMG_6297 copy.jpg contained 30,172 bytes of extraneous
data, compared with warbirdz_12327.jpg, which contained only 79 such
bytes, as determined by comparing file sizes before and after
processing with jStrip v3.3. And, actually, there are any number of
Windows compatible graphics programs that can accomplish this task
easily, but no, they do not ship with Windows. As to whether the
difference is beyond perception, I am not prepared to state such a
conclusion so definitively.

On Tue, 27 Nov 2007 22:18:28 -0700, Norm DePlume wrote:

On Tue, 27 Nov 2007 22:22:59 -0500, Andrew Kalten
In addition, IMG_6297 copy.jpg contained 30,172 bytes of extraneous
data, compared with warbirdz_12327.jpg, which contained only 79 such
bytes, as determined by comparing file sizes before and after processing
with jStrip v3.3. And, actually, there are any number of Windows
compatible graphics programs that can accomplish this task easily, but
no, they do not ship with Windows. As to whether the difference is
beyond perception, I am not prepared to state such a conclusion so
definitively.

I had no time to do a statistical analysis of the difference file,
but a quick glance using a histogram tool showed that the average
difference in luminosity was about 3-4 (out of a maximum of 255).
Since the threshold for the detection of luminosity differences
by the human eye is about one percent, this difference will just
barely exceed that level. Looking at the difference image, a vague
area of greyness is just barely discernible.

But this difference is at the lower end of the luminosity scale.
In the actual image, the differences occur in much brighter areas
where the eye is less sensitive to change and such differences are
very apt to go completely unnoticed.

So unless you are very deliberately and very intently looking for
some small difference, the images will be identical.

AK

Hello,
I agree with you that such differences are, as you said, "very apt" to
go unnoticed, although I suspect there are some who would notice. I am
reluctant to use absolutes in cyber conversations, as they are
frequently considered provocative.

On Wed, 28 Nov 2007 00:06:59 -0700, Norm DePlume wrote:

Hello,
I agree with you that such differences are, as you said, "very apt" to
go unnoticed, although I suspect there are some who would notice. I am
reluctant to use absolutes in cyber conversations, as they are
frequently considered provocative.

Human imagination can be wanton. If you present a more or less random
pattern, such as an ink blot, to the average person, he is likely to
discover many things within the meaningless form -- but such things
will have no basis in reality. That's why we always need an objective
measure.

Cyber conversations can be problematic for the same reason. If one
seeks provocation, perhaps only unconsciously, then one will tend to
discover provocation whether it exists or not.

Science is more than a white coat and a job in a laboratory. Science
is the ability to discover what really exists despite the trickery
of imagination.

AK