XPost: alt.comp.os.windows-11
On Sun, 6/1/2025 2:02 PM, knuttle wrote:
On 06/01/2025 10:54 AM, Paul wrote:
On Sun, 6/1/2025 7:19 AM, knuttle wrote:
On 06/01/2025 3:37 AM, ...winston wrote:
micky wrote:
What are resized jpeg's?
how do they get resized? I've never done this on purpose. They're on >>>>> the PC but come from a phone.
More importantly, if I delete them, is there an original somewhere else? >>>>> Using Everything to try to find a jpeg without the resized prefix
doesn't find anything. But I'm sure I have not deleted any other copy. >>>>>
I have 36 of them, and they are in my Appdata, in the directory for
MyPhoneExplorer, which I use to copy pictures from my phone to my PC, >>>>> and they are in a subdirectory named after my old phone
They have names like
000010000006_Resized_20200209_170147.jpeg
in C:\Users\mmm\AppData\Roaming\MyPhoneExplorer\samsung SM-A305F
[354863100416828]\MessageAttachments\000010000006_Resized_20200209_170147.jpeg
One has been resized twice!!!!
000010000012_Resized_Resized_20200310_163246.jpeg
What does that mean?
They seem full-height without losing clarity, so what was the original >>>>> size? But narrow, since they're from a phone,
**(my persona, mmm, is for micky, micky, micky.)
MyPhoneExplorer iirc is/was an app from a 3rd party software(FJ or something similar) but lacked resize features.
- if that still holds true, then what software are you viewing and resizing pics in Windows????
- GIMP, Paint, Photos, Irfanview, etc.
For the most part, unless the software is specifically an image-resizer set to auto-resize(none of the above fall in that category by default) resizing is user dependent.
Start with a pic that has not been resized and trace your steps to determine what you are using and doing.
Without knowing specifically what you are doing(opening, viewing, saving, etc) its close to impossible to predict where an original size image exists that is not on your phone.
The question that comes to my mind is if the user is use only resized pictures, Why are they pay hundereds of dollars for a phone that takes high resolution picutures if they only want low resolution pictures.
It does not make any difference the size of the image, the phone screen has relatively low resolution.
"The sensor in the main camera of Samsung's latest flagship smartphone,
the Galaxy S20 Ultra, is a prime example for both these trends. At 1/1.33"
it's one of the currently largest (only the 1/1.28" chip in the
Huawei P40 Pro is bigger) and a whopping 108MP resolution allows for
pixel binning and all sorts of computational imaging wizardry to produce >> 12MP high-quality default output."
And then when you get the file on the computer, it's 1600x1200 :-/
Maybe the sensor was 1600x1200 the whole time ? o.O
I expect some of the technical details are a fabrication.
*******
You can tell they don't want you to know what is going on in there.
When anyone has to resort to this. And there isn't a call with a
mode selector, to just "do it" in a library.
https://stackoverflow.com/questions/67997981/android-how-to-get-raw-image-data-from-camera
On my cheap digital camera, the resolution in the images captured,
seems to match the sensor description for the camera. There isn't
a large difference between the two to make you suspect shenanigans.
Paul
While it may not be visually apparent, anytime you change the size of the native image that was captured by the photodetector chip, you change the quality of the image. If you reduce the size of the native image some pixel are thrown out to make
the smaller image. If you increase the size of the native image, the quality is reduced as the software averages the native pixel, to fill in for the missing pixel that were created when the image was made bigger.
This is similar to a discussion I had with my brother in regard to digital zoom verse optical zoom. With digital zoom, as above, the native pixel in a small part to the native images averaged to create the new pixel required to make the larger
zoomed image.
With an optical zoom the image captured by the photo cell is the image enlarged by the lens. There is no reduction in the quality by pixel averaging to create the zoomed image.
When you count picture elements (pixels or pels), a marketing
person might count at the Bayer filter pattern level. In the
sample illustration at the top, a 2x2 group of elements allows
one RGB tuple to be derived.
https://en.wikipedia.org/wiki/Bayer_filter
On Electronic News Gathering (ENG) shoulder mount cameras,
there is a lense, and three separate sensor chips. One chip
handles red colors, one handles green colors, one handles blue colors.
If each sensor had 1024x768 elements, the output is 1024x768 RGB pixels.
Whereas camera sensors use a single sensor and the Bayer layout. They
do this, because the thickness of the phone, does not allow the same
optical element sequence (Prisms, reflectors, whatnot) that an ENG
camera would have, to take one lens image and send a copy to
three sensors.
That means, if you started with a SmartPhone 108 million "elements",
the picture resolution could be 27 million "pixels". I would
hope if shooting a picture, there should be some capability
to get all 27 million of them, not 2MB or as little as
660 thousand of them. That's off by more than an order of magnitude.
Yes, throwing away pixels is throwing away information,
information that does not come back. Interpolation does
not magically add information back. Mathematically, the
extra info added via interpolation, does not aid any further
processing, to make the picture look better. If I interpolate
by a factor of sixteen, or sixty four, or ten twenty four,
I'm not winning by doing that. So once a pixel is thrown
away to make an image smaller, it isn't coming back.
If interpolation was magic, we would do it a million times
and see galaxies far far away. The extra galactic information
would appear by magic, in the midst of the million extra
elements generated by interpolation.
Paul
--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)