| Author | Thread |
|
|
03/26/2006 10:24:41 AM · #1 |
Is it allowed to use unsharp mask in basic editing? I see in the rules that masking is strictly not allowed and I'm not sure how the "unsharp mask" tool works. Does it in the process use masking?
This is the tool I have always used in basic editing. Hope it was okay... :)
::are |
|
|
|
03/26/2006 10:27:24 AM · #2 |
It's fine, the "mask" that you aren't allowed to use in basic editing refers to a technique that applies to selected areas of the picture.
USM is a filter that applies to the whole image.
The "mask" part of that refers to the original process from which this digital equivalent comes. |
|
|
|
03/26/2006 10:42:38 AM · #3 |
Originally posted by Are_62:
Is it allowed to use unsharp mask in basic editing? I see in the rules that masking is strictly not allowed and I'm not sure how the "unsharp mask" tool works. Does it in the process use masking?
|
"Unsharp mask" is a term that originates from an old "wet" darkroom technique used for sharping. That terminology really does not apply here but relates the software back to traditional photography.
It is not a mask in the sense of the way the term is used here anymore. Here masks are used to modify the visibility of a layer and are not allowed in basic editing since they are applied selectively to the image.
Message edited by author 2006-03-26 10:46:13.
|
|
|
|
03/26/2006 11:05:28 AM · #4 |
Thanks - that helps.
On a similar note: I have been using an 'action' before that I downloaded somewhere. It sharps very nice at the click of a button. But I realized that this action uses tools that aren't allowed in basic editing mode. This is why I all of a sudden weren't sure about the unsharp mask. But it makes sense.
I attached a picture of the sharpening action (called 10D finisher - made for the Canon 10D but works well regardless). I have used this before in basic editing without thinking about what the action actually does. To me ( a beginner then), it appeared as a tool that sharpened the entire image since I never actually looked at all the steps inside the action.
I'm not using this anymore now of course
Message edited by author 2006-03-26 11:07:02. |
|
|
|
03/26/2006 11:20:14 AM · #5 |
Originally posted by Are_62:
|
Looks like the action does edge sharpening. It is selectively chosing the edges that looks to be its main problem. Can't tell if it uses illegal modes or not.
What is interesting about it is that it is intended to sharpen edges without sharpening noise. Filters are allowed to reduce noise and sharpen in basic rules to achieve both these goals, but they cannot be applied selectively on an image to do it together. Funny how these things are, isn't it?
Message edited by author 2006-03-26 11:21:07.
|
|
|
|
03/26/2006 11:30:21 AM · #6 |
| 10D finisher is a great sharpening action, but I'm 99.9% sure it would result in a DQ if validation was required in a basic challenge. It's edge selection sequence makes it pretty obvious. Watch it work sometime, it's pretty! |
|
|
|
03/26/2006 12:10:39 PM · #7 |
The first thing that makes that action illegal under basic is it duplicates the data layer several times. So, yup it's illegal as illegal gets for basic editting.
|
|
|
|
03/26/2006 12:15:16 PM · #8 |
| regardless, where did you get this action? i'd like to play with it a bit i think. |
|
|
|
03/26/2006 01:17:47 PM · #9 |
I did a search and it doesn't look like it's available anymore. It is a fantastic action however and I use all the time (except basic challenges).
But I'm absolutely certain that it was freeware. It's bundlet with two warming actions as well.
If you give me your e-mail address, I'll send it to you.
::are |
|
|
|
03/26/2006 01:38:23 PM · #10 |
Originally posted by fotomann_forever:
The first thing that makes that action illegal under basic is it duplicates the data layer several times. So, yup it's illegal as illegal gets for basic editting. |
I never really understand this site. Unsharp mask creates multiple layers as well. It generates a blurred version which is used as a mask; basically it creates a new image/layer and then combines it selectively with the original image.
If you were to do this by hand, it would be illegal in basic. Or would it not be, because it is explicitly allowed, but you're just not using the built-in command?
Likewise, an image was recently disqualified because the artist decided to use an alternative method for a built-in command that would have been allowed. Why are built-in commands allowed to violate the rules? If somebody were to write custom plug-ins that designed to remove sensor dust from their pictures, unique to their own camera (who knows, find an area that matches color and intensity near the known corrupted area and do the equivalent of a clone) would this be allowed?
I'm sure the response will be "that's too difficult for anybody to ever do" like when it was suggested that somebody might manipulate EXIF data. |
|
|
|
03/26/2006 11:51:10 PM · #11 |
It's kinda the honors system on the EXIF editing scene.
I can't really comment on most of what you said, but I will say that each camera actually already does have it's own unique profile.
Each pixel on the sensor is very slightly different in it's actual sensitivity and there are generally a small handful of dead or hot pixels kicking around on there. These are corrected at the factory by use of something called a pixel map. They set up a series of adjustments that get applied to each pixel every time you take a picture so it LOOKS like you have a well balanced sensor.
As to whether people will modify in-camera firmware to accomplish minor editing that would be easy to do in photoshop in 90% of cases, and would probably be somewhat overlooked in a basic editing rules contest... Yeah, I think it's pretty unlikely.
So far, even the Firmware modifications that have come from "external" sources have been very minimal, consisting of altering of single bits within data files to turn certain features on or off.
That's a far cry from writing sensor dust removal algorithms.
Probably easier to just clean the sensor? |
|
|
|
03/27/2006 12:05:14 AM · #12 |
Originally posted by eschelar:
It's kinda the honors system on the EXIF editing scene.
I can't really comment on most of what you said, but I will say that each camera actually already does have it's own unique profile.
Each pixel on the sensor is very slightly different in it's actual sensitivity and there are generally a small handful of dead or hot pixels kicking around on there. These are corrected at the factory by use of something called a pixel map. They set up a series of adjustments that get applied to each pixel every time you take a picture so it LOOKS like you have a well balanced sensor.
As to whether people will modify in-camera firmware to accomplish minor editing that would be easy to do in photoshop in 90% of cases, and would probably be somewhat overlooked in a basic editing rules contest... Yeah, I think it's pretty unlikely.
So far, even the Firmware modifications that have come from "external" sources have been very minimal, consisting of altering of single bits within data files to turn certain features on or off.
That's a far cry from writing sensor dust removal algorithms.
Probably easier to just clean the sensor? |
By plug-in I meant writing something that interfaces with GIMP or Photoshop. I don't actually know what an 'action' is but it seems like a macro. If you were to write something a layer below that does the same things, but as a plug-in, and not a macro, my understanding is that this could be allowed.
Something that detects abnormality and then applies filter around that area could be applied to the full image or just around the abnormality and it would have the same effect. Selective Gaussian blur is somewhat related in that it applies a Gaussian blur selectively based on a threshold differntial. Likewise, something could look for edges and apply sharpening around them, or noise reduction elsewise. In an extreme case, it could look for the known abnormality for your camera (my raw conversion software actually comes with a customizable map that will deal with dead/hot spots in a smart manner--which I believe would be held to be legal if it's part of the conversion process, but possibly not if it's applied afterwards? Where this line is drawn I'm not actually clear on either, as there ought to be a single point where raw image becomes non-raw images.) |
|
|
|
03/27/2006 09:48:52 AM · #13 |
Perhaps it would be worthwile to query the SC directly on that one. Kirbic usually seems to be the man on these sorts of questions, but submit it as a general question ticket and see what comes up.
It's not so much of an issue for me because I both don't have the photoshop skills and I try to do the best I can in-camera for the most part with my basic editing challenges.
When it says in-camera, I take it to mean in-camera.
Message edited by author 2006-03-27 09:49:56. |
|
|
|
03/27/2006 10:12:53 AM · #14 |
If the outcome has the effect and desire of localised editing than I feel that this moves away from the premise of a basic editing challenge, however this is one of those grey areas which ultimately need a judgement from the SC's.
Originally posted by m:
Originally posted by eschelar:
It's kinda the honors system on the EXIF editing scene.
I can't really comment on most of what you said, but I will say that each camera actually already does have it's own unique profile.
Each pixel on the sensor is very slightly different in it's actual sensitivity and there are generally a small handful of dead or hot pixels kicking around on there. These are corrected at the factory by use of something called a pixel map. They set up a series of adjustments that get applied to each pixel every time you take a picture so it LOOKS like you have a well balanced sensor.
As to whether people will modify in-camera firmware to accomplish minor editing that would be easy to do in photoshop in 90% of cases, and would probably be somewhat overlooked in a basic editing rules contest... Yeah, I think it's pretty unlikely.
So far, even the Firmware modifications that have come from "external" sources have been very minimal, consisting of altering of single bits within data files to turn certain features on or off.
That's a far cry from writing sensor dust removal algorithms.
Probably easier to just clean the sensor? |
By plug-in I meant writing something that interfaces with GIMP or Photoshop. I don't actually know what an 'action' is but it seems like a macro. If you were to write something a layer below that does the same things, but as a plug-in, and not a macro, my understanding is that this could be allowed.
Something that detects abnormality and then applies filter around that area could be applied to the full image or just around the abnormality and it would have the same effect. Selective Gaussian blur is somewhat related in that it applies a Gaussian blur selectively based on a threshold differntial. Likewise, something could look for edges and apply sharpening around them, or noise reduction elsewise. In an extreme case, it could look for the known abnormality for your camera (my raw conversion software actually comes with a customizable map that will deal with dead/hot spots in a smart manner--which I believe would be held to be legal if it's part of the conversion process, but possibly not if it's applied afterwards? Where this line is drawn I'm not actually clear on either, as there ought to be a single point where raw image becomes non-raw images.) |
|
|
Home -
Challenges -
Community -
League -
Photos -
Cameras -
Lenses -
Learn -
Help -
Terms of Use -
Privacy -
Top ^
DPChallenge, and website content and design, Copyright © 2001-2025 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Current Server Time: 12/29/2025 03:16:48 AM EST.
