Reading mode: Light Dark

jaggy contour in picture

Go to last post
A

Amadeus21

Senior Member
Messages
1,361
Reaction score
511
Location
DE
Today I prepared a picture for sendig it to have it printed. As I had a closer look at the details (stored out of SPP in normal and double size) I saw following jaggy contour at the tree:



These jaggies occur from time to time - mostly in those cases, when I really do not want them... They are, too, to be seen in the single size picture and make the best photos worse.
By what are they caused? Are here the Foveon photosites to be seen?

--
Johannes

http://www.huegelheim.de/weitzel/SD_10/
http://www.huegelheim.de/weitzel/SD_10-2
 
Amadeus21 said:
Today I prepared a picture for sendig it to have it printed. As I
had a closer look at the details (stored out of SPP in normal and
double size) I saw following jaggy contour at the tree:



These jaggies occur from time to time - mostly in those cases, when
I really do not want them... They are, too, to be seen in the
single size picture and make the best photos worse.
By what are they caused? Are here the Foveon photosites to be seen?

--
Johannes

http://www.huegelheim.de/weitzel/SD_10/
http://www.huegelheim.de/weitzel/SD_10-2
Click to expand...
You over sharpened before interpolation and used the wrong interpolation method. Use Bicubic softer. Actually there is no need to interpolate since the printer will do that for you for the print size you desire. You'll ge much better results if you just send them the original image.

regards,
Larry

--
http://www.fredmiranda.com/hosting/showgallery.php?ppuser=235&cat=500
 
Johannes, I have a feeling this is a blown highlight.

Maybe you can do some good in SPP by lowering the highlights?

Kind regards,
Clive
 
"Jaggies" are caused when a diagonal straight line is represented by a series of squares be they pixels or squares on a chess board. It's worse on Sigma since the Foveon chip doesn't smear the image like a Bayer sensor does to interpolate the color at each pixel location. If the SD14 has more pixels but the same size sensor this problem will be reduced.
td
--
http://www.pbase.com/tucsondave
 
Hi Larry -
LMC54 said:
Actually there is no need
to interpolate since the printer will do that for you for the print
size you desire. You'll ge much better results if you just send
them the original image.
Click to expand...
The printer will do the interpolation to enlarge the image as required, but I've seen a number of posts that recommend doing the resizing yourself and not trusting the printer (I just had a print made on a Kodak 1400 dye sub printer and could see jaggies in fine objects). And in this situation where a print is to be made by a lab, there could be a surprise when the print is delivered.

The better approach would be to resize the image using any of the common interpolation methods along with appropriate sharpening and once the results are acceptable, send that file for printing.
LMC54 said:
Amadeus21 said:
Today I prepared a picture for sendig it to have it printed. As I
had a closer look at the details (stored out of SPP in normal and
double size) I saw following jaggy contour at the tree:



These jaggies occur from time to time - mostly in those cases, when
I really do not want them... They are, too, to be seen in the
single size picture and make the best photos worse.
By what are they caused? Are here the Foveon photosites to be seen?

--
Johannes

http://www.huegelheim.de/weitzel/SD_10/
http://www.huegelheim.de/weitzel/SD_10-2
Click to expand...
You over sharpened before interpolation and used the wrong
interpolation method. Use Bicubic softer. Actually there is no need
to interpolate since the printer will do that for you for the print
size you desire. You'll ge much better results if you just send
them the original image.

regards,
Larry

--
http://www.fredmiranda.com/hosting/showgallery.php?ppuser=235&cat=500
Click to expand...
--
John P. Sabo
[email protected]

'Any sufficiently advanced technology is indistinguishable from magic.'
  • Arthur C. Clarke -
 
JPSabo said:
LMC54 said:
Actually there is no need
to interpolate since the printer will do that for you for the print
size you desire. You'll ge much better results if you just send
them the original image.
Click to expand...
The printer will do the interpolation to enlarge the image as
required, but I've seen a number of posts that recommend doing the
resizing yourself and not trusting the printer (I just had a print
made on a Kodak 1400 dye sub printer and could see jaggies in fine
objects). And in this situation where a print is to be made by a
lab, there could be a surprise when the print is delivered.

The better approach would be to resize the image using any of the
common interpolation methods along with appropriate sharpening and
once the results are acceptable, send that file for printing.
LMC54 said:
Amadeus21 said:
Today I prepared a picture for sendig it to have it printed. As I
had a closer look at the details (stored out of SPP in normal and
double size) I saw following jaggy contour at the tree:



These jaggies occur from time to time - mostly in those cases, when
I really do not want them... They are, too, to be seen in the
single size picture and make the best photos worse.
By what are they caused? Are here the Foveon photosites to be seen?

--
Johannes

http://www.huegelheim.de/weitzel/SD_10/
http://www.huegelheim.de/weitzel/SD_10-2
Click to expand...
You over sharpened before interpolation and used the wrong
interpolation method. Use Bicubic softer. Actually there is no need
to interpolate since the printer will do that for you for the print
size you desire. You'll ge much better results if you just send
them the original image.

regards,
Larry

--
http://www.fredmiranda.com/hosting/showgallery.php?ppuser=235&cat=500
Click to expand...
--
John P. Sabo
[email protected]

'Any sufficiently advanced technology is indistinguishable from
magic.'
  • Arthur C. Clarke -
Click to expand...
Yes your right. I would advise to make sure your printer is very good and uses the best RIPs. Ask the printer what RIP they use.
regards,
Larry

--
http://www.fredmiranda.com/hosting/showgallery.php?ppuser=235&cat=500
 
Thanks to you all und your suggestions!

It's really already in SPP when the picture is processed. Here you have an original view, I exported the picture as double size for printing purposes - what doubled the jaggies, too...



Thanks again.

It is such a nice picture, more exact, for me important, because I wanted to enlarge it for an exhibition. But with these "stairs" ...

--
Johannes

http://www.huegelheim.de/weitzel/SD_10/
http://www.huegelheim.de/weitzel/SD_10-2
 
Hello Johannes,
Amadeus21 said:
It is such a nice picture, more exact, for me important, because I
wanted to enlarge it for an exhibition. But with these "stairs" ...
Click to expand...
I would try and set "Schärfe" to a lower value (like -3). At "0" setting, SPP does some sharpening already which enhances your jaggies.

Next, I would save the photo as it is and remove what remains of the stairs with Photoshop (or any other photo editor), just like they were dust spots. Should be a matter of a few minutes to make them completely invisible.

Greetings, Max
 
There may not be a problem here...I don't think the magnifying glass tool is doing a quality interpolation since it's made for quick inspection. Try saving the file to a TIFF, same size and then take a look at that area.

If you can post the TIFF file to site, that might be of interest for testing some interpolation methods.

Also, what size print are you planning and what printer will produce the print?

--
John P. Sabo
[email protected]

'Any sufficiently advanced technology is indistinguishable from magic.'
  • Arthur C. Clarke -
 
Yes, that's what I was trying to give a subtil & hoflich comment to help.

When there is a significant overexposure, the anti-aliasing doesn't work as well as normally.

I think you have two good suggestions from persons here, for managing the situation:
  • adjust in SPP so that the highlight isn't so overloaded, first. You can do the best you can here without affecting other parts of the image. Thus the suggestion to adjust the Highlight control, besides others.
  • I would listen very carefully to what Larry says about painting in anti-aliasing. It's getting clear he's quite an expert in such painting.
Kind regards, and good fortune,
Clive
 
.. agree with this explanation and de prediction concerning the SD 14 by Tucsondave2. "Jaggies"or "Stair Stepping" cannot be eliminated, only made less visible by lowering the sharpening and/or contrast.
It's always possible to see this phenomene even in small prints (A3 size).

cheers, Paul
http://www.paulparislesimages.com
 
Hi Larry,

What Rip have you found to be more acceptable? I have a Epson 7600 without a Rip. I have thought about one but the $$$ is more than I can afford.

Hermann
 
Hermann said:
Hi Larry,
What Rip have you found to be more acceptable? I have a Epson 7600
without a Rip. I have thought about one but the $$$ is more than I
can afford.

Hermann
Click to expand...
I find lots of RIPs good Firey is very good. I own and use the Espon 9800 with Qimage and it's not a postscript RIP but does well and renders the aliasing very good. The OP was asking about his alaising problems after interpolating. If his using SPP to interpolate for printing I wouldn't recommend that or he'll have those stairstep messes.
regards,
Larry

--
http://www.fredmiranda.com/hosting/showgallery.php?ppuser=235&cat=500
 
I hope this is not taken the wrong way but digital does not have the resolving power of film (Da) This is an extreme blowup for digital, I tell people if you want it big fill the frame move in or zoom, also it appears your focus is on the leafs in the foreground trying to sharpen the out of focus highlight is only going to make it worse
just my thoughts from your sample, I like your Idea I would try it again!

Len
--
SD9
 
Again to you all: Thanks for the helpfull and hoflichen :-) words and advice (I'm somewhat stirred up (?) by the sympathy...)

The picture is rather important for me, for I want to use it in an exhibition context. The problem with this X3F is typical for me/the SD 10: The picture is made with tripoid. But may be, I forgot to darken the > ¿?:-((( viewfinder during exposure and the AF does not do its duty in that low light situation. (I'm of the opinion, that the viewfinder is a catastrophy for professional use. Finding the focus in lowlight is a game of chance.)

If I see it right for the whole pict, it is rather difficult to set the highlights slider to low because of the other highlight areas, and sharpening may be important because a lot of things in that picture are not satisfying sharp.

In the middle of the picture there is a very bright area at the floor of the woods. It is about to be break out.

Here I got the idea for the first time, to "develope" the picture two times in SPP: One with good look in the dark areas and one with correct colours in the highlights - and to compose the 2 hereafter via Photoshop.

Here the overview:



To play with my problems, here is the X3F (Please respect the copyright...):

http://www.huegelheim.de/weitzel/IMG02009.X3F

Thanks to all!

--
Johannes

http://www.huegelheim.de/weitzel/SD_10/
http://www.huegelheim.de/weitzel/SD_10-2
 
I think your main problem is that the backlighting is too bright and you have blown highlights. The only way to fix this in spp is to lower the exposure or fill light or both. I would do whatever it takes to get the tif or jpg out of spp without those blown highlights, not only on the tree edges but on the ferns as well. After that go to ps and use some of the tricks Larry and others are suggesting.

This is a good example of why using bracketing is a good idea. If you did bracket and you do have the same shot at a lower exposure it is easier to bring the light up from the dark than vise versa and there are many ways to blend two or more images to extend dynamic range.

Good luck,

Mike
 
Hi Johannes,

It was very helpful for you to include the X3F, so we can see what you are up against.

And it could be I discovered something that would help, and which also might help with the 'politics', which often seems to occur around problem-solving with Foveon.

I think your idea of superimposing two different developments is a pretty intelligent one, and that may help with a number of things you are trying to do.

The hard jagged edge in upper left is however a problem. I have noticed these many times on overexposed Sigma highlights, and wondered a bit. And they have seemed very hard to remove. Maybe I have just found the reason for that.

If you magnify this section, you can see a color rainbow in hard edges - not just a white against black. I think this color edging is what prevents SPP or Photoshop from anti-aliasing properly.

In fact, I found that if I take the color away from those hot spots, then a normal gaussian blur will nicely antialias the jagged edges.

To limit the area of doing this remove-color-then-blur action, we would like to use a mask, and my level of Photoshop doesn't apparently allow enough control to do this, though I think a normal one must. So I can't show it to you fully fixed, but I'm sure the words are clear.

Where does this color come from? Well, I could offer some handwaving about certain optics and semiconductors, but let's just say it's a characteristic when the exposure has far too much highlight.

Why don't we know of this before? Well, again, a complicated subject. Let's just hope a new camera opens a new field and manner of discussion for everyone.

Best for your success, and weekend in any case. And I hope this helps.

Clive
Amadeus21 said:
Again to you all: Thanks for the helpfull and hoflichen :-) words
and advice (I'm somewhat stirred up (?) by the sympathy...)

The picture is rather important for me, for I want to use it in an
exhibition context. The problem with this X3F is typical for me/the
SD 10: The picture is made with tripoid. But may be, I forgot to
darken the > ¿?:-((( viewfinder during exposure and the AF does not
do its duty in that low light situation. (I'm of the opinion, that
the viewfinder is a catastrophy for professional use. Finding the
focus in lowlight is a game of chance.)

If I see it right for the whole pict, it is rather difficult to set
the highlights slider to low because of the other highlight areas,
and sharpening may be important because a lot of things in that
picture are not satisfying sharp.

In the middle of the picture there is a very bright area at the
floor of the woods. It is about to be break out.
Here I got the idea for the first time, to "develope" the picture
two times in SPP: One with good look in the dark areas and one with
correct colours in the highlights - and to compose the 2 hereafter
via Photoshop.

Here the overview:



To play with my problems, here is the X3F (Please respect the
copyright...):

http://www.huegelheim.de/weitzel/IMG02009.X3F

Thanks to all!

--
Johannes

http://www.huegelheim.de/weitzel/SD_10/
http://www.huegelheim.de/weitzel/SD_10-2
Click to expand...
 
Downloaded, opened with PS, it's normal with such scale of detail. And our Larry has very good experience in resize tool as PS plugin to deal with such jaggy problem.



Rgds,
Hung
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top