source camera identification

[Viking]

i just found out that they can identify the source camera from a picture
uploaded to the internet and placed on web pages or uploaded by others.

is there a counter tool for privacy on Windows that changes a photo so that
its not trackable by the umnique pixel deformities of the camera?

https://www.sciencedirect.com/science/article/pii/S0925231216303319
https://ieeexplore.ieee.org/document/1530330

[Savageduck]

On Jun 7, 2019, Viking wrote
(in article <qdfddh$50s$1...@gioia.aioe.org>):

...and it has taken you 14 years to find this, and why so paranoid?

--
Regards,
Savageduck

[David B.]

This article should help you, 'Viking':-

https://us.norton.com/internetsecurity-how-to-how-to-remove-gps-and-other-metadata-locations-from-photos.html

Please let us know!

[Carlos E.R.]

He is not talking of metadata.

--
Cheers, Carlos.

[David B.]

Oops! I made the same mistake too. :-(

Well spotted, Carlos!

--
David B.
Devon, UK

[Savageduck]

On Jun 8, 2019, Carlos E.R. wrote
(in article <df8ssf-...@Telcontar.valinor>):

Check the dates of the articles.

--
Regards,
Savageduck

[Ken Hart]

I want to see them try to identify my camera. I'm reasonably certain my
photos don't contain any metadata, or unique (to the camera) pixel
deformities.

--
Ken Hart
kwh...@frontier.com

[Viking]

David B. responded at time Sat, 8 Jun 2019 08:10:09 +0100 :

> https://us.norton.com/internetsecurity-how-to-how-to-remove-gps-and-other-metadata-locations-from-photos.html
>
> Please let us know!

the question is how best to change the image to make details less traceable

Tiny details in photos identify your unique phone
https://www.futurity.org/smartphones-cameras-prnu-1634712-2/

[Viking]

Ken Hart responded at time Sat, 8 Jun 2019 12:08:01 -0400 :

> I want to see them try to identify my camera. I'm reasonably certain my
> photos don't contain any metadata, or unique (to the camera) pixel
> deformities.

the question isn't whether they can because certainly they can.

Determining Digital Image Origin Using Sensor Imperfections
http://ia.binghamton.edu/publication/FridrichPDF/EI5685-29_Forensic.pdf

the question is how to change the image so that its harder for them.

[Savageduck]

On Jun 8, 2019, Ken Hart wrote
(in article <qdgmh0$5ab$1...@gioia.aioe.org>):

You sneaky analogist you. ;-)

--
Regards,
Savageduck

[Savageduck]

On Jun 8, 2019, Viking wrote
(in article <qdgnhm$9vb$1...@gioia.aioe.org>):

You seem to have an affinity for finding old articles, this one is 2 years
old, and the original study it is drawn from is 10 years old. The article in
your next post, and your earlier posts are equally ancient.

My question for you remains; Why this concern about having the source camera
of a digital image identified?

--
Regards,
Savageduck

[David B.]

Very interesting. I've never heard of this before. Something for 'bad
guys' to be wary of, that's for sure! ;-)

Facebook removes metadata from photos before publishing them on-line. I
don't know if they also remove photo-response non-uniformity (PRNU)
evidence too.

Thanks for advising, 'Viking'. :-)

[nospam]

In article <qdgmh0$5ab$1...@gioia.aioe.org>, Ken Hart

<kwh...@frontier.com> wrote:

>
> I want to see them try to identify my camera. I'm reasonably certain my
> photos don't contain any metadata, or unique (to the camera) pixel
> deformities.

it's incredibly easy to identify your camera model, even though there
aren't any pixels at all, deformed or otherwise.

[Ken Hart]

And you would do that... how?
Other than looking at my Facebook page or eBay past purchases.

--
Ken Hart
kwh...@frontier.com

[Viking]

Savageduck responded at time Sat, 08 Jun 2019 10:07:40 -0700 :

> You sneaky analogist you. ;-)

i don't know about analog cameras but for digital cameras the sensor has
imperfections that need to be smoothed out for better digital photo privacy
online.

alls I'm asking is if proficiant digital owners know which techniques in
software like irfanview or photoshop or gimp will reduce the number of
identifying bits in an image without totallly ruining the image.

[nospam]

In article <qdh2oa$1q2r$1...@gioia.aioe.org>, Ken Hart

<kwh...@frontier.com> wrote:

> >>
> >> I want to see them try to identify my camera. I'm reasonably certain my
> >> photos don't contain any metadata, or unique (to the camera) pixel
> >> deformities.
> >
> > it's incredibly easy to identify your camera model, even though there
> > aren't any pixels at all, deformed or otherwise.
> >
>
> And you would do that... how?

easily.

> Other than looking at my Facebook page or eBay past purchases.

no need.

[nospam]

In article <qdh2vu$1s5h$1...@gioia.aioe.org>, Viking

<gone_a...@vikingwewillgo.net> wrote:

>
> i don't know about analog cameras but for digital cameras the sensor has
> imperfections that need to be smoothed out for better digital photo privacy
> online.

no.

> alls I'm asking is if proficiant digital owners know which techniques in
> software like irfanview or photoshop or gimp will reduce the number of
> identifying bits in an image without totallly ruining the image.

some do.

[Viking]

Ken Hart responded at time Sat, 8 Jun 2019 15:36:43 -0400 :

> And you would do that... how?
> Other than looking at my Facebook page or eBay past purchases.

this is only about digial sensors fingerprinting not analog
where its a simple question of basic digital privacy online

Digital Camera Identification from Sensor Pattern Noise
http://www.ws.binghamton.edu/fridrich/Research/double.pdf

are any skillful photogreaphers here who can suggest how to smooth the
unique fingerprinted imperfections in digital sensor images without ruining
it?

[nospam]

In article <qdh3cu$1ttm$1...@gioia.aioe.org>, Viking

<gone_a...@vikingwewillgo.net> wrote:

> Digital Camera Identification from Sensor Pattern Noise
> http://www.ws.binghamton.edu/fridrich/Research/double.pdf
>
> are any skillful photogreaphers here who can suggest how to smooth the
> unique fingerprinted imperfections in digital sensor images without ruining
> it?

no need. read your own link for further information.

[Viking]

nospam responded at time Sat, 08 Jun 2019 15:52:22 -0400 :

> no need. read your own link for further information.

they test two fixes where the second fix is in need of tools

http://www.ws.binghamton.edu/fridrich/Research/double.pdf

"The second problem we now investigate is whether it is possible
to make an arbitrary image look as if it was taken by a specific camera.
Again, having access to the reference pattern or the camera makes
this indeed possible. We denoised 20 Canon G2 pictures and added to them
the reference pattern from Canon S40. We increased the amplitude of
the added noise, till we reached a correlation that was higher
than the correlation the image previously had with the Canon G2
reference pattern. The peak signal to noise ratio (PSNR) for the
forged images was on average 37.5dB and the images were visually
indistinguishable from originals. The forgeries still had higher
correlations with the Canon G2 reference pattern than expected
from different camera images, but this correlation could be
eliminated using some of the techniques mentioned above. We also
tried to plant the noise extracted from a single Canon S40
image using the denoising filter [12] into a denoised
Canon G2 image. The correlation of the noise extracted
from the resulting image with the Canon S40 reference
pattern was usually within the range of the typical correlation
values achieved by other Canon S40 images. This kind of malicious
processing requires multiplying the added noise with a perceptual
mask [11] to avoid creating visible artifacts. Overall, we conclude
that malicious manipulation that will fool the identification
algorithm is, indeed, possible if the attacker possesses enough
skill in signal processing. We note that it is unlikely that
there exists a numerical identification characteristic computed from
digital images that could not be compromised by a sufficiently
sophisticated opponent. All previously proposed techniques based on
defective pixels [3] [6] or image features [4], are certainly vulnerable
to malicious attacks as well. "

what I'm asking is for a windoze tool that does this on purpose?

[Viking]

David B. responded at time Sat, 8 Jun 2019 18:39:07 +0100 :

> Thanks for advising, 'Viking'. :-)

alls I know is there is techniques in the conclusion of this reported work
http://www.ws.binghamton.edu/fridrich/Research/double.pdf

"Overall, we conclude that malicious manipulation that will fool
the identification algorithm is, indeed, possible if the attacker
possesses enough skill in signal processing. We note that it is
unlikely that there exists a numerical identification characteristic
computed from digital images that could not be compromised by a
sufficiently sophisticated opponent. All previously proposed
techniques based on defective pixels [3] [6] or image features [4],

are certainly vulnerable to malicious attacks as well. Since the
identification technique requires proper synchronization,
geometrical operations, such as cropping, resizing, rotation,
digital zoom, cause desynchronization and prevent correct camera
identification. an informed attacker can suppress the pattern noise
by dark frame subtraction and flat-fielding. However, images are
not typically stored in raw formats and are only available as
TIFF/JPEG, which means they are already processed in the
camera (Eq. (2) in Section II). As a result, it is in general
not possible to perform flat fielding correctly from a TIFF/JPEG
image. A simpler way to remove the pattern noise PC, well-known to
researchers working in robust watermarking, is as follows. The
attacker can arrange for ñC = 0 for any image p taken with C by solving the
equation corr(p+áPC, PC) = 0 with respect to á and taking p+áPC as the
forged image. An uninformed attacker could attempt to remove the pattern
noise by applying the same denoising filter. While this,
indeed, decreases the correlation value with the correct pattern
approximately by a factor of two, in most cases correct
identification will still be possible. However, repetitive
application of the filter or more aggressive denoising filters
will likely prevent correct identification. The easiest way to
prevent a simple detection of the reference pattern is
desynchronization, such as slight rotation, possibly combined with other
processing that might include resizing, cropping, and filtering.
Probably the simplest activemeasure that the photographer can use
to complicate image identification later is to take images using a
continuous digital zoom, a feature that many consumer digital
cameras have today."

do you know of winsoze tools specific to clean up digital fingerprints?

[Savageduck]

On Jun 8, 2019, Viking wrote
(in article <qdh2vu$1s5h$1...@gioia.aioe.org>):

My question remains. Why do you NEED to do this?

Are you trying to clean up Digimark watermarks, or other embedded ID other
than EXIF data?

As I have pointed out, all of your source material is very dated, so why the
urgency in anonymizing these digital images.

--
Regards,
Savageduck

[David B.]

Are there some specific tracks which you are trying to cover?

If it's really important, I'll ask around for you ... but I need to know
*WHY* you want such a tool.

[Viking]

On Sat, 8 Jun 2019 22:29:21 +0100, David B. wrote:

> If it's really important, I'll ask around for you ... but I need to know
> *WHY* you want such a tool.

are you still beating your wife?

your conspiracy questions are frustrating because they are absurd.
its just basic privacy and not an al quaida conspiracy on the usenet.

nobody even understood what I was asking it seems where so far I haven't
found software that is custom made to spoof or forge the image sensor
imperfections.

does anyone on this group know anythinug about digital photos or not?

[Viking]

On Sat, 8 Jun 2019 22:29:21 +0100, David B. wrote:

> If it's really important, I'll ask around for you ... but I need to know
> *WHY* you want such a tool.

are you still beating your wife?

your conspiracy questions are frustrating because they are absurd.
its just basic privacy and not an al quaida conspiracy on the usenet.

only one or two people even understood what I was asking it seems where so

far I haven't found software that is custom made to spoof or forge the
image sensor imperfections.

does anyone on this group know of software that will fool the typical
identification algorithm by either spoofing or removing identifying bits?

[nospam]

In article <qdhm1a$shm$1...@news.mixmin.net>, Viking

<gone_a...@vikingwewillgo.net> wrote:

> does anyone on this group know anythinug about digital photos or not?

many do.

they also know that you're trolling.

[Viking]

nospam responded at time Sat, 08 Jun 2019 21:20:18 -0400 :

>> does anyone on this group know anythinug about digital photos or not?
>
> many do.

none did. but I will give it time for someone who does.

since you think this question is trolling i will only answer posts that are
about this question and not about some al quaida consipriacy in their mind.

you were one of few who understood the digital photo question is to prevent
synchronization by a series of edits such as cropping, resizing, slight
rotation, repetitive denoising and taking the photo with non uniform
digital zoom.

so far I have not found any windoze software that is designed to prevent or
hinder synchronization of digital photos for camera identification or
spoofing but I can do those cropping and editing tasks in any software.

I just need to learn what those tasks are that prevent synchronization.

[Savageduck]

On Jun 8, 2019, Viking wrote
(in article <qdho6t$koo$1...@gioia.aioe.org>):

> nospam responded at time Sat, 08 Jun 2019 21:20:18 -0400 :
>
> > > does anyone on this group know anythinug about digital photos or not?
> >
> > many do.
>
> none did. but I will give it time for someone who does.

Actually when it comes to digital photos, and for some of us digital & film
photography, quite a large group of the regular posters to this NG, and the
other photo groups have years of experience, and know quite a large amount on
the subject.

>
> since you think this question is trolling i will only answer posts that are
> about this question and not about some al quaida consipriacy in their mind.

Your post, and your sudden appearance in this NG reeks of trolling. Some of
us in this NG are trying to understand the entire purpose of your query so
that we might be able to answer you question.

...and who other than you said anything about conspiracy theories, and “al
quaida”?

>
> you were one of few who understood the digital photo question is to prevent
> synchronization by a series of edits such as cropping, resizing, slight
> rotation, repetitive denoising and taking the photo with non uniform
> digital zoom.

Actually when it comes to digital photography what you are describing has
nothing to do with image capture by a digital camera, but digital editing.
Your use of “synchronization” in your question is vague and meaningless.

>
> so far I have not found any windoze software that is designed to prevent or
> hinder synchronization of digital photos for camera identification or
> spoofing

That alone should tell you something regarding the futility of your question.

> but I can do those cropping and editing tasks in any software.

Good for you. Consider that none of those cropping and editing tasks are
camera specific, and few, if any are made in camera.

>
> I just need to learn what those tasks are that prevent synchronization.

As I said, your use of “synchronization” is meaningless, and most
importantly irrelevant, especially for those of us engaged in digital, and
for some film, photography. None of us are particularly concerned that some
esoteric lab might be capable of identifying a particular source camera. Most
of us limit our interest to the EXIF metadata which is included in most of
our digital images, but can be stripped if needed.

If you actually told us what you were trying to achieve other than this
“need to learn” how to prevent this meaningless, and irrelevant
“synchronization” we might be able to provide some guidance. Otherwise,
we see little more than trollish behavior.

--
Regards,
Savageduck

[nospam]

In article <qdho6t$koo$1...@gioia.aioe.org>, Viking

<gone_a...@vikingwewillgo.net> wrote:

>
> >> does anyone on this group know anythinug about digital photos or not?
> >
> > many do.
>
> none did.

false.

> but I will give it time for someone who does.

what for?

there's plenty of information out there, including in the link *you*
provided, which you obviously do not understand.

> since you think this question is trolling i will only answer posts that are
> about this question and not about some al quaida consipriacy in their mind.

further confirming it's trolling.

> you were one of few who understood the digital photo question is to prevent
> synchronization by a series of edits such as cropping, resizing, slight
> rotation, repetitive denoising and taking the photo with non uniform
> digital zoom.

you said none knew anything above, and now you say a few do.

> so far I have not found any windoze software that is designed to prevent or
> hinder synchronization of digital photos for camera identification or
> spoofing but I can do those cropping and editing tasks in any software.

try other platforms.

> I just need to learn what those tasks are that prevent synchronization.

there's a lot you need to learn.

[Viking]

Savageduck responded at time Sat, 08 Jun 2019 21:00:16 -0700 :

> As I said, your use of "synchronization" is meaningless, and most
> importantly irrelevant

http://www.ws.binghamton.edu/fridrich/Research/double.pdf

"Since the identification technique requires proper

SYNCHRONIZATION, geometrical operations, such as cropping,
resizing, rotation, digital zoom, cause DESYNCHRONIZATION and
prevent correct camera identification."

"The easiest way to prevent a simple detection of the

reference pattern is DESYNCHRONIZATION, such as slight rotation,

possibly combined with other processing that might include
resizing, cropping, and filtering."

http://ia.binghamton.edu/publication/FridrichPDF/EI5685-29_Forensic.pdf
"Since our camera identification is essentially based on detection of a
high frequency watermark using correlation, geometrical transformations of
images, such as rotation, cropping, resizing, or fish-eye lens processing
cause DESYNCHRONIZATION and must be corrected for prior to calculating the
correlation."

"The pattern noise is a low-amplitude high-frequency ´natural watermark¡
and as such, it is well known that the easiest way to prevent its detection
is DESYNCHRONIZATION, such as slight rotation, possibly combined with other
processing that might include resizing, cropping, and JPEG compression."

"While the proposed identification technique can be used for reliable
camera identification from original images or images processed using JPEG
compression or resizing, simultaneous application of other geometrical
operations (e.g., cropping resizing, rotation) causes DESYNCHRONIZATION and
thus increases the computational complexity of pattern detection because
the detection will likely have to resort to brute force searches."

[David B.]

On 09/06/2019 02:12, Viking wrote:
> On Sat, 8 Jun 2019 22:29:21 +0100, David B. wrote:
>
>> If it's really important, I'll ask around for you ... but I need to know
>> *WHY* you want such a tool.
>
> are you still beating your wife?

How do you know that I'm married?!!!

> your conspiracy questions are frustrating because they are absurd.
> its just basic privacy and not an al quaida conspiracy on the usenet.

Not true. It's *NOT* basic privacy.

> only one or two people even understood what I was asking it seems where so
> far I haven't found software that is custom made to spoof or forge the
> image sensor imperfections.

That's _proof_ that it is *NOT* basic privacy.

> does anyone on this group know of software that will fool the typical
> identification algorithm by either spoofing or removing identifying bits?

I doubt that they would tell you even if they did!

Things are not always as they seem. ;-)

--
David B.
Devon, UK

Tekrider warning:
https://www.dropbox.com/s/6nj382qhv4wzmju/Tekrider.net%20-%20Infected%20with%20malware%20%28Sucuri%29.tiff?dl=0

[newshound]

On 08/06/2019 17:25, Viking wrote:

That is a *really* interesting paper. While it ends up getting pretty
technical, it provides a really accessible description of several
different approaches, as well as stuff about sensor technology.

[Viking]

newshound responded at time Sun, 9 Jun 2019 11:59:01 +0100 :

> That is a *really* interesting paper. While it ends up getting pretty
> technical, it provides a really accessible description of several
> different approaches, as well as stuff about sensor technology.

thank you for reading the paper where I don't know what they mean by
"denoising filter" which seems to be the key component of their camera
sensor unique identification technology.

"The pattern noise is extracted from the images using a wavelet-based
denoising filter. For each camera under investigation, we first determine
its reference pattern, which serves as a unique identification fingerprint.
This could be done using the process of flat-fielding, if we have the
camera in possession, or by averaging the noise obtained from multiple
images, which is the option taken in this paper. To identify the camera
from a given image, we consider the reference pattern noise as a
high-frequency spread spectrum watermark, whose presence in the image is
established using a correlation detector. Using this approach, we were able
to identify the correct camera out of 9 cameras without a single
misclassification for several thousand images. Furthermore, it is possible
to perform reliable identification even from images that underwent
subsequent JPEG compression and/or resizing."

do you understand this "denoising" process?
is denoising something we can try on our own to see what it does?

[Savageduck]

On Jun 9, 2019, Viking wrote
(in article <qdj545$r0o$1...@gioia.aioe.org>):

I have also read the antique paper, and understand it. Consider that for
those of us who deal with noise there is both luminance noise, and color
noise, each are different and are handled differently. Also too heavy a hand
with noise reduction is going to result in a soft, or muddy image with
reduced sharpness. There is no free ride.

Denoising, or noise reduction is available in most image processing software,
and as stand-alone, or plugin apps. most have trial editions. That goes for
Lightroom/ACR(Photoshop), Exposure X4, On1 Photo RAW, Luminar, Affinity
Photo, Capture One, DxO Optics Pro, and more. For stand-alone/plugins there
are/or were Noiseless CK, Noise Ninja, Topaz Denoise, NIK Define, Noiseware,
and more.

Ultimately most of us here don’t give a damn if the camera we use is
identifiable. Our concern is if the visible noise degrades the image quality,
and can be reduced without destroying the image. So why is it an issue for
you?

--
Regards,
Savageduck

[Viking]

Savageduck responded at time Sun, 09 Jun 2019 08:27:46 -0700 :

> Denoising, or noise reduction is available in most image processing software,

this paper surveys about 20 most common of the image denoising techniques
https://www.cse.unr.edu/~fredh/papers/conf/034-asoidt/paper.pdf

in order to solve the stated problem I will need to understand them better
where they mention wavelets which came about around 1995

the other papers also mentioned these wavelets as the denoising approach
they used so I need to better understand wavelet sparsity & multiresolution
structure.

that paper says there are two image denoising approaches
spatial filters (linear & non-linear)
transform domain filtering (data adaptive & non data adaptive)
where the paper goes into details on each

the solution will probably concentrate a better understanding of the
non-linear wavelet threshold filtering techniques using soft filtering to
prevent artifacs since thats what the original papers used.

it may be photoshop can solve the id & spoof problem but its too early to
say just yet

[nospam]

In article <qdjdmh$2n8$1...@gioia.aioe.org>, Viking

<gone_a...@vikingwewillgo.net> wrote:

>
> it may be photoshop can solve the id & spoof problem but its too early to
> say just yet

it's not too early.

[Ron C]

Clearly you do not understand the link between spread spectrum and
synchronization.
When you figure that out you should have the answer you are seeking.
--
==
Later...
Ron C
--

[Bill W]

Arlen? Is that you?

[newshound]

On 08/06/2019 17:25, Viking wrote:

> David B. responded at time Sat, 8 Jun 2019 08:10:09 +0100 :
>
>> https://us.norton.com/internetsecurity-how-to-how-to-remove-gps-and-other-metadata-locations-from-photos.html
>>
>> Please let us know!
>
> the question is how best to change the image to make details less traceable
>
> Tiny details in photos identify your unique phone
> https://www.futurity.org/smartphones-cameras-prnu-1634712-2/
>
It's all very interesting stuff technically.

Personally, I'd have thought that most photographers were interested in
techniques by which they could prove that an image that had been used
commercially had, in fact, been pirated from their post.

I can think of two sorts of people who might want to conceal the source
of an image. One might be (say) a cabinet minister who photographed a
confidential document and then leaked it. Another would be producers of
child pornography. Neither, in my view, are deserving of protection.

I suppose a whistleblower might have some reason to try to conceal their
identity. But on the other hand if an image showed signs of serious and
"smart" digital manipulation to a forensic specialist this might tend to
discredit it as a possible forgery.

I might be wrong, but I suspect that for modern cameras with rapidly
increasing pixel counts, it will be much easier to hide pixel variation
with perhaps a couple of random crops and rotations prior to
compression. Particularly for a single image, even if the camera is
already in the hands of the investigator (provided the photographer is
smart enough to avoid cloud storage).

[EllisMorgan]

This link describes the basics of wavelets and mentions their use in
image processing. It is dated 2016 and called "Understanding Wavelets -
Part 1", there are three more parts and many references to follow.
https://www.youtube.com/watch?v=QX1-xGVFqmw
.

[Incubus]

On 2019-06-08, David B. <BDo...@REMOVE.gmail.com> wrote:
> On 08/06/2019 17:25, Viking wrote:
>> David B. responded at time Sat, 8 Jun 2019 08:10:09 +0100 :
>>
>>> https://us.norton.com/internetsecurity-how-to-how-to-remove-gps-and-other-metadata-locations-from-photos.html
>>>
>>> Please let us know!
>>
>> the question is how best to change the image to make details less traceable
>>
>> Tiny details in photos identify your unique phone
>> https://www.futurity.org/smartphones-cameras-prnu-1634712-2/
>

> Very interesting. I've never heard of this before. Something for 'bad
> guys' to be wary of, that's for sure! ;-)
>
> Facebook removes metadata from photos before publishing them on-line.

Not before they have stored that metadata and shared it with the security
services.

[David B.]

That wouldn't surprise me at all ....

.... but how do you know that to be true?

Did you hear it on the Internet?

https://www.youtube.com/watch?v=34CHqptjj9E

Assuming it *IS* true, does that bother you?

[Incubus]

It's a privacy issue. It used to be the case that in order to spy on people,
tap their telephones, bug their homes etc., a warrant had to be issued. That
has been effectively circumvented by doing the same thing through devices we
connect to the Internet. It should bother anyone who thinks the government
should stay out of our business unless we are committing an offence.

[nospam]

In article <qdlvn2$tqm$1...@dont-email.me>, Incubus

<incubus...@gmail.com> wrote:

>
> It's a privacy issue. It used to be the case that in order to spy on people,
> tap their telephones, bug their homes etc., a warrant had to be issued.

a warrant is still required to tap phones or place a bug in a home.

> That
> has been effectively circumvented by doing the same thing through devices we
> connect to the Internet.

it has not.

> It should bother anyone who thinks the government
> should stay out of our business unless we are committing an offence.

government isn't the problem. it's big business, who doesn't need a
warrant to collect anything, especially when it's voluntarily given.

[David B.]

If it 'bothers' *you*, 'Incubus' then you must unplug yourself from the
Internet completely.

In my opinion, there is *NO* privacy online - computers, tablets, smart
phones and watches - they are ALL spying on you! ;-)

There's much helpful information here:-

https://en.wikipedia.org/wiki/Internet_privacy

One simply has to get used to the fact that things have changed and that
the average citizen has absolutely no clue whatsoever about Internet
.... surveillance, shall I call it? If you are 'connected', you are 'known'!

[Incubus]

On 2019-06-10, nospam <nos...@nospam.invalid> wrote:
> In article <qdlvn2$tqm$1...@dont-email.me>, Incubus
><incubus...@gmail.com> wrote:
>
>>
>> It's a privacy issue. It used to be the case that in order to spy on people,
>> tap their telephones, bug their homes etc., a warrant had to be issued.
>
> a warrant is still required to tap phones or place a bug in a home.
>
>> That
>> has been effectively circumvented by doing the same thing through devices we
>> connect to the Internet.
>
> it has not.

In the UK it has.

https://en.wikipedia.org/wiki/Investigatory_Powers_Act_2016

>> It should bother anyone who thinks the government
>> should stay out of our business unless we are committing an offence.
>
> government isn't the problem. it's big business, who doesn't need a
> warrant to collect anything, especially when it's voluntarily given.

That big business is a problem, I agree, and it is actively working with a
number of governments around the world.

[PT French]

newshound said:

> I might be wrong, but I suspect that for modern cameras with rapidly
> increasing pixel counts, it will be much easier to hide pixel variation
> with perhaps a couple of random crops and rotations prior to
> compression.

it is very interestang in it takes much more than just crops and
compression where tiny rotations seems to be a better start accoding to the
papers

the papers didnt say exactly what to do for basic privacyh online but these
are some of the steps thaty talked about

first take the picture using a digital zoom if thats possible
usually not so the first step then is to resize (but thats not enough)
then do a lossy crop
then rotate slightly (maybe 1 degree in photoshop?)
then run a set of repetitive denoise operations (which photoshop ones?)
then run filtering (maybe a slight fish eye lens in photoshop?)
only later run a jpeg compression since compression & resizing dont work
well all by itself

the result just has to look good online compared to the original where youd
do this for every picter posted online so it has to be scripted in
photoshop ahead of time.

once i figure out the exact steps I can automate it into the save menu.
https://www.dummies.com/software/adobe/photoshop/how-to-automate-a-series-of-steps-in-photoshop-cs6/
HOW TO AUTOMATE A SERIES OF STEPS IN PHOTOSHOP CS6

which denoise do you think is best?

[nospam]

In article <qdm33t$1vg$1...@gioia.aioe.org>, PT French

<ptfr...@is.invalid> wrote:

>
> which denoise do you think is best?

lightroom

[nospam]

In article <qdm1n8$c1g$1...@dont-email.me>, Incubus

<incubus...@gmail.com> wrote:

> >> It's a privacy issue. It used to be the case that in order to spy on
> >> people,
> >> tap their telephones, bug their homes etc., a warrant had to be issued.
> >
> > a warrant is still required to tap phones or place a bug in a home.
> >
> >> That
> >> has been effectively circumvented by doing the same thing through devices
> >> we
> >> connect to the Internet.
> >
> > it has not.
>
> In the UK it has.
>
> https://en.wikipedia.org/wiki/Investigatory_Powers_Act_2016

if you mean this:
€ allowed police, intelligence officers and other government
department managers (listed below) to see the Internet connection
records, as part of a targeted and filtered investigation, without a
warrant;[43]

that's not the same as a phone tap or a bug in a house.

it's equivalent to phone call logs, which has long been available
without a warrant.

it's also trivial to make it impossible for anyone to see, inside our
outside of the uk, with or without a warrant.

[PT French]

EllisMorgan said:

> This link describes the basics of wavelets and mentions their use in
> image processing. It is dated 2016 and called "Understanding Wavelets -
> Part 1", there are three more parts and many references to follow.
> https://www.youtube.com/watch?v=QX1-xGVFqmw

thank you for undersatnding the question which has nothing to do with the
date of the papers given the first haar mother wavelet was discovered in
1910 adn providng specific answers to the question instead of making up
idiotic al quaida conspiracies which anyone stupid can make up who has no
idea of anything technical about the question

the whole point is to change the image so that you dont voluntarily give
away your privacy every time you upload images on the net if its easy to
not do that just by integrating a seresis of filters into the photoshop
save

in that first video Kirthi Develeker from mathworks explained why the fft
wont work for functions localized in time and frequency where wavelets are
rapidly decaying wavelike oscillation with zero mean
morlet daubechies coiflets biorthogonal mexican hat symlets etc

that first video talked about wavaelet scale & shifting where
scale & frequency are proportional around the center frequency

the second video was better because it talked about denoising and
compression of images. from the third video which talked about wavelet
denoising and soft thresholding I realized I need to randomize the denoise
and rotation operations in photoshosp that will be tied to the save button

the fourth video talked about separating singla from noiise with fft and
wavelet analysis where wavelet cwt function looked at an earthquake energy
spread localized in time and freuqncy which isn't easy to do with fft.

using icwt they reconstructed the original time domain representation which
i think is analogous to what they did in the wavelet papers if I understood
that on my first lietening (I will need to listen to the videos many more
times though).

now i can see where wavelet comes from since its a little wave

using a big window (scale) they can pull out low frequencies (big wave) and
using a small window (scale) they can pull out the high frequency
components (little waves).

theres only one fourier transform but there are all kinds of wavelets which
is what the image uniquely contains if I don't know how to get rid of them

it seems that the most important things to do to an image as I save it is
rotation and denoise where rotation of 1 degree (randomized) is what I will
use first but I'm not yet sure which denoise filter to use.

best suggestions on the approach to tie to the photoship save button to
prevent synchronization of images posted online?

[PT French]

nospam said:

> lightroom

lightroom is good as a starting tool becuase it keeps the changes in the
catalog set of instructions without actually changing the photo where the
exported image is what is volumntarily uploaded for all the world to see

which lightroom filter is a good starting point is the next question where
it seems the graduated radial filter can handle useful image wavelet
desynchronization parameters like white balance, tint, saturation etcetera

[nospam]

In article <qdm66u$g0l$1...@gioia.aioe.org>, PT French

<ptfr...@is.invalid> wrote:

>
> > lightroom
>
> lightroom is good as a starting tool becuase it keeps the changes in the
> catalog set of instructions without actually changing the photo where the
> exported image is what is volumntarily uploaded for all the world to see

known as a non-destructive workflow.

> which lightroom filter is a good starting point is the next question where
> it seems the graduated radial filter can handle useful image wavelet
> desynchronization parameters like white balance, tint, saturation etcetera

no it's not.

[David B.]

Are you a professional photographer?

Regardless, where may readers view some of your work?

[nospam]

In article <c0xLE.185620$ID4.1...@fx06.fr7>, David B.

<BDo...@REMOVE.gmail.com> wrote:

>
> Regardless, where may readers view some of your work?

beware of the stalking request.

[Savageduck]

On Jun 10, 2019, nospam wrote
(in article<100620191348014072%nos...@nospam.invalid>):

Well there are others, some of which deliver somewhat different, if not
better results than LR. However, LR is pretty good, but for the purposes of
this thread, and the proposed workflow best is irrelevant as the result is
going to be a soft, mushy mess.

--
Regards,
Savageduck

[Savageduck]

On Jun 10, 2019, PT French wrote
(in article <qdm33t$1vg$1...@gioia.aioe.org>):

> newshound said:
>
> > I might be wrong, but I suspect that for modern cameras with rapidly
> > increasing pixel counts, it will be much easier to hide pixel variation
> > with perhaps a couple of random crops and rotations prior to
> > compression.

Holy crap!! The level of paranoia this thread has generate has reached a
level where therapy should be recommended.

>
> it is very interestang in it takes much more than just crops and
> compression where tiny rotations seems to be a better start accoding to the
> papers
>
> the papers didnt say exactly what to do for basic privacyh online but these
> are some of the steps thaty talked about
>
> first take the picture using a digital zoom if thats possible

For this idiotic source camera ID obscuring I am not going to change my
shooting habits, I shoot with both zoom, and prime lenses.

>
> usually not so the first step then is to resize (but thats not enough)
> then do a lossy crop
> then rotate slightly (maybe 1 degree in photoshop?)

All pointless.

>
> then run a set of repetitive denoise operations (which photoshop ones?)

More idiocy, as all that will result in is, severe damage to the IQ.

>
> then run filtering (maybe a slight fish eye lens in photoshop?)
> only later run a jpeg compression since compression & resizing dont work
> well all by itself

You might do that, but I use Photoshop to improve any of my images I care to
present, and what you propose is not going to do anything but damage the
image. That is not my goal

>
> the result just has to look good online compared to the original where youd
> do this for every picter posted online so it has to be scripted in
> photoshop ahead of time.

I don’t believe that any of the regular contributors to this, or any other
photography NG who use Lightroom, Photoshop, or any other good editing
software would buy into such a workflow. I certainly wouldn’t.

>
>
> once i figure out the exact steps I can automate it into the save menu.
> https://www.dummies.com/software/adobe/photoshop/how-to-automate-a-series-of-s
> teps-in-photoshop-cs6/
> HOW TO AUTOMATE A SERIES OF STEPS IN PHOTOSHOP CS6

That can be done, but the question remains; WHY?

I for one include EXIF data embedded in the great majority of my images
posted online, and that is going to ID the source camera, and lens. Many
others here do the same.

>
>
> which denoise do you think is best?

It doesn’t matter, what you have suggested is going to destroy the image as
anything resembling a quality photography. The more I see , and read in this
thread, the more it reeks of trollishness.

BTW: Are you perhaps Viking, now switched to PT French?

--
Regards,
Savageduck

[Incubus]

On 2019-06-10, nospam <nos...@nospam.invalid> wrote:

> In article <qdm1n8$c1g$1...@dont-email.me>, Incubus
><incubus...@gmail.com> wrote:
>
>> >> It's a privacy issue. It used to be the case that in order to spy on
>> >> people,
>> >> tap their telephones, bug their homes etc., a warrant had to be issued.
>> >
>> > a warrant is still required to tap phones or place a bug in a home.
>> >
>> >> That
>> >> has been effectively circumvented by doing the same thing through devices
>> >> we
>> >> connect to the Internet.
>> >
>> > it has not.
>>
>> In the UK it has.
>>
>> https://en.wikipedia.org/wiki/Investigatory_Powers_Act_2016
>
> if you mean this:
> € allowed police, intelligence officers and other government
> department managers (listed below) to see the Internet connection
> records, as part of a targeted and filtered investigation, without a
> warrant;[43]
>
> that's not the same as a phone tap or a bug in a house.

"introduced new powers, and restated existing ones, for UK intelligence
agencies and law enforcement to carry out targeted interception of
communications, bulk collection of communications data, and bulk interception
of communications"

"permitted the police and intelligence agencies to carry out targeted
equipment interference, that is, hacking into computers or devices to access
their data"

[David B.]

Are you one of Shadow's henchmen? He's a liar too.

Talking of waves, I found and like, this painting:-

https://en.wikipedia.org/wiki/The_Ninth_Wave#/media/File:Hovhannes_Aivazovsky_-_The_Ninth_Wave_-_Google_Art_Project.jpg