Re the second, you enhance (or ruin) an image by manipulating individual pixels. Don't see how you can do it otherwise!
There is not an artist of value who doesn't understand the physical characteristics of their paints, and no sculpture of value who doesn't understand the molecular structure of their stone, and no woodturner of value who doesn't know the same about their wood... and no digital photographer of value who doesn't take loving care of each of his pixels!:D
But a pixel is just three numbers. It is not sharp or blurred. It is neither beautiful nor ugly. It has no subtleties. These qualities only apply to the set of pixels we call an image.
But a pixel is just three numbers. It is not sharp or blurred. It is neither beautiful nor ugly. It has no subtleties. These qualities only apply to the set of pixels we call an image.
Exactly! Sharp or blurred, beautiful or ugly, depend both on the data in the pixel and aesthetic judgments. An image is exactly both science and aesthetics! Like any other art!
Boy, I'm with Richard on B&W conversion. If that doesn't lose information, I don't know what does. As Richard points out, otherwise you could reconstruct a color image automatically from B&W.
And I guess I'm maybe never going to get this concept. Is there something about resolution? About being able to look at the image at any magnification? Suppose we had infinite resolution, dynamic range, and depth of field. Suppose we had spectral color. Suppose we had some way to display this all. Would this make for perfect pixel quality?
Maybe I'm in need of some examples of good vs bad pixel quality. Ansel Adams vs Henri Cartier Bresson. I assume Adams wins hands down even though some of us may strongly prefer HCB. Which of your own images have pixel quality you are most proud of? Which lack it? What about that last post attempt of mine? What about the blown highlights and plugged shadows examples? Let's have some examples to ground this discussion!
Maybe I'm in need of some examples of good vs bad pixel quality. Ansel Adams vs Henri Cartier Bresson. I assume Adams wins hands down even though some of us may strongly prefer HCB. Which of your own images have pixel quality you are most proud of? Which lack it? What about that last post attempt of mine? What about the blown highlights and plugged shadows examples? Let's have some examples to ground this discussion!
This is an interesting discussion, but I don't think it belongs in the Landscape forum.
Steph: feel free to split this into two threads if you're feeling energetic.
Hello Richard and rutt
@ rutt: Really still don't get it? Well, help is abundantly available to
you with a little searching on the internet.
I have given a clear example with curves - posterisation, where lightening
shadows is achieved at the expense of reducing bit depth (physical pixel
quality) and causing undesirable image appearance (aesthetic pixel
quality).
"Posterization occurs when an image's apparent bit depth has been decreased so much that it has a visual impact... Any process which 'stretches' the histogram has the potential to cause posterization. Stretching can be caused by techniques such as levels and curves in Photoshop, or by converting an image from one color space into another as part of color management. The best way to ward off posterization is to keep any histogram manipulation to a minimum." - cambridgeincolour.com
"...posterization can, sometimes, be significantly reduced or even
eliminated by reducing the amount of image editing. For example, reducing
the amount of Curves or Hue/Saturation used might help reduce or eliminate
a posterization problem. Another possibility is to use a tool that doesn't
spread the tones apart (e.g., use Brightness/Contrast instead of Curves).
By reducing the amount of editing, or using a tool that does not spread the
tones apart, the distance between the tones will be less and the
posterization problem may disappear." - Ron Bigelow
"I just wanted to give you a sense of how incredibly important these image pixels are, especially since we’re going to be changing their
appearance, for better or for worse, when working on
our images in Photoshop." - Steve Patterson, PhotoshopEssentials.com
"One of the realities of digital image processing is that every time you make a change to a digital image you slightly degrade it." -
LuminousLandscape.com
"Image quality depends on pixel quality." - me!
@ rutt, re your mention of resolution: resolution is related to a physical
quantity - the numer of pixels in a sensor or display. It is not related to
pixel quality, though it certainly is to image quality.
@ Richard & rutt, re color information: there are fully saturated red,
green and blue filters arranged on the photodetector pixels of the sensor.
Using the information these produce via the photodetectors, the sensor chip
records three tone profiles for a population of pixels, one for each filter. (Further processing of the information occurs later in the in-camera processing.)
The tone profiles are not color information, they are information about
tone brightness on a scale from black to white. The Bayer filters do not create color. No color comes out of the camera. What come out of the camera are tonal maps. This is information that display software can use to translate into a color space and produce a color image, but it is not color, it is tone variations/brightness ranges. Digital images produced in color are governed by the same concepts of sampling, quantization, spatial resolution, bit depth, and dynamic range that apply to their grayscale counterparts. However, instead of a single brightness map of gray levels or shades of gray, color images have pixels that are quantized using three independent brightness components, producing three brightness maps, one for each of the primary colors.
"In general, all processing operations that are performed on grayscale
images can be extended to color images by applying the algorithms to each
color channel separately, then combining the channels. Color images are
constructed of three individual channels (red, green, and blue) that have
their own "gray" scales consisting of varying brightness levels for each
color. The levels are combined within each pixel to represent the final
image." - hamamatsu.com
What a display does to give you a color image is translate these
"grayscale" profiles into a color space using software algorithms.
The information used by a display to produce color is only lost if the channels data is lost. Color images can be reconstituted from b&w images if the channels information is retained. If the channels information is condensed and locked, as when the RGB luminance channel is used to make a b&w conversion, the individual channels information is lost.
Well, gentlemen ..........it's over to you:D
I trust the relocation of this thread did not risk the interest of dgrinners who really do want to communicate about the science and art of digital photography. That would be beaurocracy at its worst!
Thanks to all participants in this exciting experience!
I would like my landscape images to be returned to the forum I took a great deal of trouble to post them to, and in which some people there showed some interest, please.
Comments
But a pixel is just three numbers. It is not sharp or blurred. It is neither beautiful nor ugly. It has no subtleties. These qualities only apply to the set of pixels we call an image.
Exactly! Sharp or blurred, beautiful or ugly, depend both on the data in the pixel and aesthetic judgments. An image is exactly both science and aesthetics! Like any other art!
http://www.behance.net/brosepix
And I guess I'm maybe never going to get this concept. Is there something about resolution? About being able to look at the image at any magnification? Suppose we had infinite resolution, dynamic range, and depth of field. Suppose we had spectral color. Suppose we had some way to display this all. Would this make for perfect pixel quality?
This is an interesting discussion, but I don't think it belongs in the Landscape forum.
Steph: feel free to split this into two threads if you're feeling energetic.
Photos that don't suck / 365 / Film & Lomography
@ rutt: Really still don't get it? Well, help is abundantly available to
you with a little searching on the internet.
I have given a clear example with curves - posterisation, where lightening
shadows is achieved at the expense of reducing bit depth (physical pixel
quality) and causing undesirable image appearance (aesthetic pixel
quality).
"Posterization occurs when an image's apparent bit depth has been decreased so much that it has a visual impact... Any process which 'stretches' the histogram has the potential to cause posterization. Stretching can be caused by techniques such as levels and curves in Photoshop, or by converting an image from one color space into another as part of color management. The best way to ward off posterization is to keep any histogram manipulation to a minimum." - cambridgeincolour.com
"...posterization can, sometimes, be significantly reduced or even
eliminated by reducing the amount of image editing. For example, reducing
the amount of Curves or Hue/Saturation used might help reduce or eliminate
a posterization problem. Another possibility is to use a tool that doesn't
spread the tones apart (e.g., use Brightness/Contrast instead of Curves).
By reducing the amount of editing, or using a tool that does not spread the
tones apart, the distance between the tones will be less and the
posterization problem may disappear." - Ron Bigelow
"I just wanted to give you a sense of how incredibly important these
image pixels are, especially since we’re going to be changing their
appearance, for better or for worse, when working on
our images in Photoshop." - Steve Patterson, PhotoshopEssentials.com
"One of the realities of digital image processing is that every time you
make a change to a digital image you slightly degrade it." -
LuminousLandscape.com
"Image quality depends on pixel quality." - me!
@ rutt, re your mention of resolution: resolution is related to a physical
quantity - the numer of pixels in a sensor or display. It is not related to
pixel quality, though it certainly is to image quality.
@ Richard & rutt, re color information: there are fully saturated red,
green and blue filters arranged on the photodetector pixels of the sensor.
Using the information these produce via the photodetectors, the sensor chip
records three tone profiles for a population of pixels, one for each filter. (Further processing of the information occurs later in the in-camera processing.)
The tone profiles are not color information, they are information about
tone brightness on a scale from black to white. The Bayer filters do not create color. No color comes out of the camera. What come out of the camera are tonal maps. This is information that display software can use to translate into a color space and produce a color image, but it is not color, it is tone variations/brightness ranges. Digital images produced in color are governed by the same concepts of sampling, quantization, spatial resolution, bit depth, and dynamic range that apply to their grayscale counterparts. However, instead of a single brightness map of gray levels or shades of gray, color images have pixels that are quantized using three independent brightness components, producing three brightness maps, one for each of the primary colors.
"In general, all processing operations that are performed on grayscale
images can be extended to color images by applying the algorithms to each
color channel separately, then combining the channels. Color images are
constructed of three individual channels (red, green, and blue) that have
their own "gray" scales consisting of varying brightness levels for each
color. The levels are combined within each pixel to represent the final
image." - hamamatsu.com
What a display does to give you a color image is translate these
"grayscale" profiles into a color space using software algorithms.
The information used by a display to produce color is only lost if the channels data is lost. Color images can be reconstituted from b&w images if the channels information is retained. If the channels information is condensed and locked, as when the RGB luminance channel is used to make a b&w conversion, the individual channels information is lost.
Well, gentlemen ..........it's over to you:D
I trust the relocation of this thread did not risk the interest of dgrinners who really do want to communicate about the science and art of digital photography. That would be beaurocracy at its worst!
Thanks to all participants in this exciting experience!
http://www.behance.net/brosepix
I would like my landscape images to be returned to the forum I took a great deal of trouble to post them to, and in which some people there showed some interest, please.
Thank you.
Neil
http://www.behance.net/brosepix
Moderator of the Technique Forum and Finishing School on Dgrin
Thanks and best wishes!
http://www.behance.net/brosepix