«

On 16/09/2012 17:08, HerHusband wrote:
> Peter,
>
>> I have some scanned JPEGs from some old photos,
>> and over time they have faded and become speckled.
>> This has happened only to a few.
>> Why has this happened?
>
> Assuming this isn't a hoax, the only reason a JPG photo (or any digital
> file) would change over time is if the media they are stored on is failing
> (or a virus has corrupted the files).

The problem is that the fault shown is not characteristic of a corrupt
JPEG stream - that usually affects the decoding of the rest of the file
in a gross and unmistakable way. This salt and pepper noise was encoded
into the JPEG file and has been faithfully reproduced!
>
> The first thing I would do is scan your drive for surface errors. One or
> two errors isn't a big deal, they will be mapped so they won't be used
> again. But when you start getting bad sectors, it usually escalates fairly
> quickly, so I would rescan the drive periodically until you can replace it
> with a new drive (which I would do as soon as possible).
>
> Next, I would scan the computer for viruses and spyware. The free
> MalwareBytes anti-malware works good for this.

It is just about possible that a virus or trojan is doing it but all the
ones I have ever encountered merely replace the unfortunate file with
the same number of zeros.

--
Regards,
Martin Brown

Peter Jason <> wrote:
> On Thu, 13 Sep 2012 17:18:16 -0400, Alan Browne
> <> wrote:
>
>> On 2012.09.11 21:55 , Peter Jason wrote:
>>> I have some scanned JPEGs from some old photos,
>>> and over time they have faded and become speckled.
>>>
>>> This has happened only to a few.
>>>
>>> Why has this happened?
>>
>> Were the photos repeatedly opened and re-saved?
>
> They have passed thru Microsoft "Digital Image
> Suite 9" first as tiff files and then converted to
> jpgs (after which the Suite didnt work any more),
> but some are OK.
>
> An example of a decayed photo of 1988 scanned in
> from negative color film. Note the fine white
> snow throught the picture.
> http://imageshack.us/a/img832/1747/1988a1decay.jpg

Seen this before. I don't know if it was permanent or due to the viewer.
Too few occurrences.

Greg

On 17/09/2012 01:32, gregz wrote:
> Peter Jason <> wrote:
>> On Thu, 13 Sep 2012 17:18:16 -0400, Alan Browne
>> <> wrote:
>>
>>> On 2012.09.11 21:55 , Peter Jason wrote:
>>>> I have some scanned JPEGs from some old photos,
>>>> and over time they have faded and become speckled.
>>>>
>>>> This has happened only to a few.
>>>>
>>>> Why has this happened?
>>>
>>> Were the photos repeatedly opened and re-saved?
>>
>> They have passed thru Microsoft "Digital Image
>> Suite 9" first as tiff files and then converted to
>> jpgs (after which the Suite didnt work any more),
>> but some are OK.
>>
>> An example of a decayed photo of 1988 scanned in
>> from negative color film. Note the fine white
>> snow throught the picture.
>> http://imageshack.us/a/img832/1747/1988a1decay.jpg
>
> Seen this before. I don't know if it was permanent or due to the viewer.

I don't know of any JPEG decoder that would make such a complete mess of
a valid image file. I do know plenty of "equalise histogram" functions
that would do this or even worse to an underexposed image.

> Too few occurrences.

The OP has been unhelpful to the point of totally misleading in
describing his workflow since the image he shows as a sample of the
problem *WAS NOT* saved by the package that he said he used!

--
Regards,
Martin Brown

On Mon, 17 Sep 2012 08:02:59 +0100, Martin Brown
<|||newspam|||@nezumi.demon.co.uk> wrote:

>On 17/09/2012 01:32, gregz wrote:
>> Peter Jason <> wrote:
>>> On Thu, 13 Sep 2012 17:18:16 -0400, Alan Browne
>>> <> wrote:
>>>
>>>> On 2012.09.11 21:55 , Peter Jason wrote:
>>>>> I have some scanned JPEGs from some old photos,
>>>>> and over time they have faded and become speckled.
>>>>>
>>>>> This has happened only to a few.
>>>>>
>>>>> Why has this happened?
>>>>
>>>> Were the photos repeatedly opened and re-saved?
>>>
>>> They have passed thru Microsoft "Digital Image
>>> Suite 9" first as tiff files and then converted to
>>> jpgs (after which the Suite didnt work any more),
>>> but some are OK.
>>>
>>> An example of a decayed photo of 1988 scanned in
>>> from negative color film. Note the fine white
>>> snow throught the picture.
>>> http://imageshack.us/a/img832/1747/1988a1decay.jpg
>>
>> Seen this before. I don't know if it was permanent or due to the viewer.
>
>I don't know of any JPEG decoder that would make such a complete mess of
>a valid image file. I do know plenty of "equalise histogram" functions
>that would do this or even worse to an underexposed image.
>
>> Too few occurrences.
>
>The OP has been unhelpful to the point of totally misleading in
>describing his workflow since the image he shows as a sample of the
>problem *WAS NOT* saved by the package that he said he used!

I have forgotten all the steps because I scanned
the film with a Minolta 35mm slide/negative
scanner as tiff files which I reduced later to
jpeg. This was in approx 2003.

I'm thinking of doing it all again with a faster
scanner.

Is there any proper software for databaseing the
3000 photos I have with appropriate keywords, like
the Microsoft "Digital Image Suite 9"? My
workflow at the moment is:

1/ Dump photos from the camera into a folder.
2/ Copy all these into another (working) folder
called "Cropped6x4, PShopped, Culled".
3/ Crop 6 x 4, adjust levels, and sharpness,
annotate on the photo itself, and save as jpeg.

Is this OK? Notice I keep the originals.

On 17/09/2012 23:14, Peter Jason wrote:
> On Mon, 17 Sep 2012 08:02:59 +0100, Martin Brown
> <|||newspam|||@nezumi.demon.co.uk> wrote:
>
>> On 17/09/2012 01:32, gregz wrote:
>>> Peter Jason <> wrote:

>>>> An example of a decayed photo of 1988 scanned in
>>>> from negative color film. Note the fine white
>>>> snow throught the picture.
>>>> http://imageshack.us/a/img832/1747/1988a1decay.jpg
>>>
>>> Seen this before. I don't know if it was permanent or due to the viewer.
>>
>> I don't know of any JPEG decoder that would make such a complete mess of
>> a valid image file. I do know plenty of "equalise histogram" functions
>> that would do this or even worse to an underexposed image.
>>
>>> Too few occurrences.
>>
>> The OP has been unhelpful to the point of totally misleading in
>> describing his workflow since the image he shows as a sample of the
>> problem *WAS NOT* saved by the package that he said he used!
>
> I have forgotten all the steps because I scanned
> the film with a Minolta 35mm slide/negative
> scanner as tiff files which I reduced later to
> jpeg. This was in approx 2003.
>
> I'm thinking of doing it all again with a faster
> scanner.

If you still have the original TIFF scan of the sample image you put up
then it should be possible to work out what went wrong and where. My
money is still on a seriously underexposed original image.

Also the image was rather small in size for a slide or negative scan.
>
> Is there any proper software for databaseing the
> 3000 photos I have with appropriate keywords, like
> the Microsoft "Digital Image Suite 9"? My

No idea. I don't particularly like database software.

> workflow at the moment is:
>
> 1/ Dump photos from the camera into a folder.
> 2/ Copy all these into another (working) folder
> called "Cropped6x4, PShopped, Culled".
> 3/ Crop 6 x 4, adjust levels, and sharpness,
> annotate on the photo itself, and save as jpeg.
>
> Is this OK? Notice I keep the originals.

Probably. It depends how well you do the image adjustments.
I would not bother cropping unless I was intending to reprint YMMV.

I generally scan stuff as a batch and then adjust later.

--
Regards,
Martin Brown

On Wed, 12 Sep 2012 11:55:24 +1000, Peter Jason <> wrote:

>Why has this happened?

The photons are wearing out - you'll have to order some new ones from
CERN.

On Sat, 15 Sep 2012 17:16:18 +1000, Peter Jason <> wrote:

>An example of a decayed photo of 1988 scanned in
>from negative color film. Note the fine white
>snow throught the picture.
>http://imageshack.us/a/img832/1747/1988a1decay.jpg

It's given that woman in the background choppers like a horse.
Why, that would cost a fortune any other way in dental charges.