Are we using the term, "Resolution" accurately?

[koppelman]

I have searched the archives for the answer to my question but have been unsuccessful....

I have become confused by the use of the term, "Resolution" in the field of digital photography.

Resolution is defined as the ability to distinguish individual components of an image. Intuitively, I would suppose that the ability to distinguish individual components (to perceive detail) in an image is a function of the number of dots comprising the image per unit area. (I will use the terms "dots" and "pixels" interchangeably. I do this out of ignorance)

Maximum resolution of monitors is given in total numbers of pixels in a vertical and horizontal dimension, and not the number of pixels per unit area. Therefore a larger monitor has more gross number of pixels than a smaller one, and by inference, the larger monitor has a higher resolution than the smaller one.

But can we say the the larger monitor has a better ability to distinguish individual components of an image than a smaller one?

Thanks
koppelman

[kgosden]

Since the resolution of a digital image is limited by the initial number of pixels available at the time of capture I see no problem with defining a camera's capbilities as the sensor's inherent x and y array. However, that is it's maximum resolution only if every other factor involved is able to meet or exceed it. Even in your monitor example it is only, as you said, a maximum resolution. Display a 640x480 image on each monitor and that is the actual resolution at that time.

[timmciglobal]

Resolution is ... in some degree... pixels. The problem is now that we've got 10+ mega pixels the resolution needs to be modified to "actual distinguishable resolution" which is in other reviews done in various ways by resolution charts or in lens reviews on a body.

For example, 100 mega pixels doesn't give more "resolution" unless the optics can support delivery of light accurately enough to give those 100 mega pixels meaning.

The ISO noise/sensor characteristics should be taken into account too. As 100 mega pixels of noisy pixels actually delivers less total "Resolution" then 25 not as noisy in some cases.


Tim

[John_Reed]

I have searched the archives for the answer to my question but have been unsuccessful....

I have become confused by the use of the term, "Resolution" in the field of digital photography.

Resolution is defined as the ability to distinguish individual components of an image. Intuitively, I would suppose that the ability to distinguish individual components (to perceive detail) in an image is a function of the number of dots comprising the image per unit area. (I will use the terms "dots" and "pixels" interchangeably. I do this out of ignorance)

Maximum resolution of monitors is given in total numbers of pixels in a vertical and horizontal dimension, and not the number of pixels per unit area. Therefore a larger monitor has more gross number of pixels than a smaller one, and by inference, the larger monitor has a higher resolution than the smaller one.
I think you're in error there. "Resolution," to my knowledge, is measured in pixels/inch, or dots/inch (which are not necessarily interchangeable; more later). For example, if you take two monitors, each one with a resolution of 72 pixels/inch, but one having 1200 X 800 pixels total, and the other having 1600 X 1200 pixels total, and then displayed a 640 X 480 pixel image on each, the measured size of that image on each display would be identical, since they both display at 72 pixels/inch.


But can we say the the larger monitor has a better ability to distinguish individual components of an image than a smaller one?

Thanks
koppelman
See above. You'll be able to see more of the image on the larger monitor, but you won't be able to distinguish details any better, if both have the same resolution.

One thing that muddies the water is the difference between printer dots and image pixels. I have a printer that prints at 1440 dots/inch. What happens to the pixels of an image I send to the printer, where the image resolution is 300 pixels/inch? The answer lies in the printer software, that works to resolve the incoming pixels and generate a string of dots to be printed, per line of the image. In fact, to match the pixel exactly, the printer would like to print 1440/300, or 4.8 dots/pixel. Since it can't do fractional dots, it then has to dither the output, using 5 dots for some pixels, 4 for others, so that the average still comes out correctly at 4.8.

I hope that was helpful?

[zmikers]

If I am not mistaken, Which I usually am:p, Pixel is the amount of information that your camera can process and DPI or dots per inch is the amount of information the your computer monitor or printer can output. A combination of the two would be the "resolution" of the image you are viewing on your monitor or printing.

[John_Reed]

If I am not mistaken, Which I usually am:p, Pixel is the amount of information that your camera can process and DPI or dots per inch is the amount of information the your computer monitor or printer can output. A combination of the two would be the "resolution" of the image you are viewing on your monitor or printing.Seems like this issue is more confusing than it should be.

If you start with the CCD sensor array in your camera, the thing that collects light when the shutter is open, it's a two-dimensional array of sensors. If it's a 7MP camera, like my TZ3, the array, for a 4:3 aspect ratio, has 3072 X 2304 sensors. The incoming light is actually filtered, and certain sensors are designated "Green" (50%), "Red" (25%), and "Blue (25%), in a "Bayer" matrix. Now, the camera, after the image is captured by the sensors, has to transform the collection of sensor data into pixels. The camera's output is generated as pixels, one for each of the positions in the array, and each issued pixel comprised of a Green, a Red, and a Blue component. So the camera actually must interpolate those sparse sensors to create output pixels. That is, 50% of the output green components are interpolated, and 75% of the output red and blue components are interpolated. But there's the identity of a "pixel," it has three components of information, more than a single "dot."

If the "pixel" is to be displayed on an RGB monitor, that monitor also deals with 3-levels per pixel, so it's proper to refer to monitor elements as "pixels." Printers, on the other hand, at least inkjets, process monochrome dots, so what they receive, and how they're specified, is in "dots/inch."

What I said about the Bayer sensor, which is the basis of most digital cameras, doesn't hold for the Foveon sensor, which actually collects an entire pixel at each photo site on its surface, and thus no interpolation is necessary within the camera to create a output file from the captured pixels.

And also, RAW capture takes what each little sensor site captures and delivers it directly as output, not to become properly known as a "pixel" until it is processed within the receiving computer.

I don't think a combination of the two makes much sense. If you want to measure "resolution" at any point along the way, you could measure it in "sensors/mm" on the CCD surface, "pixels/in" on the display, "dots/in" on the printer. They're all different measurements, but each one expresses "resolution."

[zmikers]

Seems like this issue is more confusing than it should be.
And your post isn't any more confusing???????:D Sorry, just takin' the piss!

[zmikers]

I don't think a combination of the two makes much sense. If you want to measure "resolution" at any point along the way, you could measure it in "sensors/mm" on the CCD surface, "pixels/in" on the display, "dots/in" on the printer. They're all different measurements, but each one expresses "resolution."

So, if finding a resolution at any point along the way is measuring the above measurements, wouldn't the resolution of a finished product be a combination of the three? The resolution of a finished photo is defined by the resolution of the sensor and the resolution it was printed in. Is this not correct?

[John_Reed]

So, if finding a resolution at any point along the way is measuring the above measurements, wouldn't the resolution of a finished product be a combination of the three? The resolution of a finished photo is defined by the resolution of the sensor and the resolution it was printed in. Is this not correct?But I've always heard resolution quoted only at any particular step. For example, what's the printer resolution? Answer: 1440 dots/inch. What was the pixel resolution of the image you printed? Answer: 160 pixels/inch. I can give you an extreme example, the 2MP image which I "blew up" using Genuine Fractals to a 22MP image, which I rendered at 150 pixels/inch. This was printed by a printer with a dot resolution of 1200 dots/inch. The original photo resolution had nothing to do with the printing parameters.

I'm not trying to argue, just to try to explain and clarify; if I'm having the opposite effect, I apologize.

[koppelman]

If you want to measure "resolution" at any point along the way, you could measure it in "sensors/mm" on the CCD surface, "pixels/in" on the display, "dots/in" on the printer. They're all different measurements, but each one expresses "resolution."

I think you have confirmed what I thought was the case. Resolution needs to be measured as a function of unit area as you have put forth in your reply. A monitor with a maximum resolution of 72 pixels per inch has a higher resolution than one with a maximum resolution of 50 pixels per inch.

So then perhaps my beef should be with the marketing people who peddle monitors by quoting total pixels instead of pixels per unit area. These are the same bunch of folks who always use the adjectives, "better" or "faster" but never provide a point of reference.

[zmikers]

But I've always heard resolution quoted only at any particular step. For example, what's the printer resolution? Answer: 1440 dots/inch. What was the pixel resolution of the image you printed? Answer: 160 pixels/inch. I can give you an extreme example, the 2MP image which I "blew up" using Genuine Fractals to a 22MP image, which I rendered at 150 pixels/inch. This was printed by a printer with a dot resolution of 1200 dots/inch. The original photo resolution had nothing to do with the printing parameters.

I'm not trying to argue, just to try to explain and clarify; if I'm having the opposite effect, I apologize.

Mate, I'm not trying to argue either:D. Sometimes its hard to understand someone's tone while reading. Maybe I should have said this before, I'm not entirely sure about this so my remarks were not argumentative but I was trying to understand better. Here's my question, for clarity and not an argumentative purpose. If you have a printer that has a resolution of X and you print two images from that same printer. One image was taken from a 3mp camera and the second from a 12mp camera. If the printer is the same do these images have the same resolution?

[Honest Gaza]

Stolen from Photoshop CS2 for Dummies.

Image Resolution :
Image resolution is the size of your image's individual pixels when you print.

Camera Resolution :
Digital cameras capture each image in a specific number of pixels.
Take a look at the number of pixels that your camera records for width and for the height. Multiply the numbers together, divide by one million, and round off the result. That's the megapixel rating for the camera.

Monitor Resolution :
Monitor resolution determines how many pixels are visible onscreen.

Printer Resolution :
Unlike the three preceding terms, printer resolution doesn't involve pixels. Rather, a printer's resolution tells you how many tiny droplets of ink are sprayed on the paper. Remember that it takes several droplets to produce a single image pixel - you certainly don't need an image resolution anywhere close to the printer's resolution.

Resolving Image Resolution :
Image resolution is nothing more than an instruction to a printing device about how large to produce each pixel. Onscreen, when working in Photoshop, your image has no resolution at all. An image that's 3000 pixels wide and 2000 pixels tall looks and acts exactly the same in Photoshop whether you've got the image resolution at 300 ppi or 72 ppi. Same number of pixels, right ?

You can always check (or change) a picture's resolution by using the Photoshop Image->Image Size command. The Image Size dialog box has two separate but related sets of information about your image. At the top, you see information about the actual image itself, in the Pixel Dimensions area. Below, in the Document Size area, you see instructions for a printing device - that "size" pertains only to printing and has no impact on what you do in Photoshop.