0

I've just made an image post-processing to increase a dynamic range.

Here is an original input image enter image description here

Here is an output image enter image description here

How can I calculate the dynamic range from a photo? In especially, how do we calculate how much dynamic range has increased?

update

modify option for detail enter image description here

original-resized enter image description here

output-resized enter image description here

Community
  • 1
greeting
  • 23
  • 4
  • 2
    The dynamic range hasn't increased at all., the amount od visible detail in any given portion of it has changed... not really for the better imho, but the dynamic range is identical [assuming both are actually untagged 8-bit RGB]. Your dynamic range is limited by that definition. – Tetsujin Jul 06 '18 at 16:41
  • To what end do you need this info? Are you just bragging to your friend that you increased DR by X% or is there something else you're trying to figure out? – OnBreak. Jul 06 '18 at 16:42
  • 2
    Also, it looks like you simply took up the shadows and blacks. Your after image lacks contrast to me and the noodles on the right lost detail. Generally speaking, you don't want to lose detail when attempting hdr. – OnBreak. Jul 06 '18 at 16:45
  • @Corey I'm just wondering what does dynamic range 50dB, 100dB and 150dB mean. from here https://kintronics.com/ip-camera-test-review-wide-dynamic-range-sony-vs-samsung-vs-hikvision/ – greeting Jul 06 '18 at 16:50
  • They're not actually widening the range, strictly speaking they're compressing it, averaging so more of the image is within 'visual tolerance' for a point & shoot. Btw, while I was messing... I'd have gone more towards https://i.stack.imgur.com/5VhwA.jpg for that shot [low qual cos it's a small jpg] – Tetsujin Jul 06 '18 at 16:55
  • @Tetsujin I'd love to upload the original image, but their image size is over 2MB. So I've changed to JPG from BMP. so you can feel like that. – greeting Jul 06 '18 at 17:00
  • 4
    Possible duplicate of What is dynamic range and how is it important in photography? : it seems that what you need is a good definition :) – Olivier Jul 06 '18 at 17:17
  • Personally, I think the image original is great. If anything, dodge the chef just a tad, as if you had used a reflector to give him some extra light. Burn the background and lower right corner just a bit and boom. Done. That, or get crazy and motion blue the chef - which would also add something to the photo. Either way, great job on the initial photo. – OnBreak. Jul 06 '18 at 17:27
  • 1
    I read your article and am still confused. The article compares cameras in their ability to capture a scene with a wide range of exposure values. This is very different than the HDR process or simply editing to move the histogram around. Can you try to rephrase your question? – OnBreak. Jul 06 '18 at 17:29
  • @Corey It's not just editing to move the histogram around(it's compressing). As you can see that updated image, even if you have a limited dynamic range camera(original image), but you can good dynamic range image(output image).you can save the money. and I'm not sure. but it also occur in HDR camera processing. – greeting Jul 06 '18 at 17:57
  • -1 It's entirely unclear to me what you are asking. Explain the used process. – null Jul 06 '18 at 18:45
  • I'm pretty sure the process is irrelevant, the image is 8-bit, the dynamic range is... 8-bit. – Tetsujin Jul 06 '18 at 19:19
  • What problem are you trying solve? – mattdm Jul 06 '18 at 22:29
  • It's all still very unclear. Are you talking about the range of your image or a camera. You mention saving money, so are you comparing different cameras?? For what it's worth, I think you may have a misunderstanding of what dynamic range is. Voting to close as a dupe. If not, please try to reframe the question and give more info on the problem you are trying to solve. – OnBreak. Jul 07 '18 at 06:45
  • @Corey , My question was "How can I calculate the dynamic range from a photo?". and if we can't, then Is there any method with 2 photos(input and output), and especially, I want to know what does 50,100, and 150dB mean of WDR – greeting Jul 07 '18 at 10:09
  • Range is a feature of the film or sensor. Once you have an image, that image is composed of pixel values - it doesn't have range. – OnBreak. Jul 07 '18 at 15:50
  • 1
    3.01 dB equals one Ev equals one step in the zone system equals log-base-2 of the contrast ratio numerator. 150dB would be about 30Ev. If you say you can measure the DR of a scene by way of an image (I'm sure CNN could do this emperically or subject-inferentially) then you could compute dB from scene metrics. The answers below are correct, though. The information you've presented does not allow for a measurement of DR. The tonal range / contrast ratio expansion or compression could be calculated but that's not what you've asked for. – PhotoScientist Jul 09 '18 at 22:53
  • @Corey, I have to disagree with you. An image with min 0 and max 255 has a contrast ratio of 256:1 or a dynamic range of 8 stops/ Ev. If the image is edited or captured such that it's min is 63 and max 255 it would have a contrast ratio of 4:1 and a dynamic range of 7.5 Ev. Note that the scene may have had tonalities which extended beyond the maximum value and also that those 7.5 Ev of image tones represent just 3Ev of sensor analog binning. Therefore it would be more correct to state the although images have dynamic range, it is not related to scene or sensor DR in a meaningful way. – PhotoScientist Jul 09 '18 at 23:01
  • @PhotoScientist agreed. In too few words I tried to make the point that you better stated. My main thought is this: comparing output images, especially edited images, is not a good representation of DR. The ability of a sensor to capture a high range, if the main goal, then the file needs to be compared to reality...not an image that's been edited. – OnBreak. Jul 10 '18 at 01:27
  • True @Corey what OP has asked is not possible. It would be interesting to explore the use of HDR software as a photometry tool but that's not what is being asked about here. It would be interesting to see the problem OP is trying to solve. – PhotoScientist Jul 10 '18 at 02:34

2 Answers2

3

The dynamic range hasn't increased.

You are looking to extrapolate data that doesn't exist. The dynamic range is dictated by the limitations of the sensor capturing the image. If an area is too dark (or too bright) it zeros out. If you were to just look at the basic RGB of a portion of an image it would look like this:

0,0,0    0,0,0    0,0,0            Visualization
                                       ■ ■ ■
0,0,0    0,0,0    0,0,0         <----- ■ ■ ■
                                       ■ ■ ■
0,0,0    0,0,0    0,0,0

At this point there is nothing you can do to recover more data.

Your technique just raises the shadows.

The dynamic range can only shrink, and that happens once you start adjusting the black/white points or converting from a RAW format.

Hairy Dresden
  • 456
  • 2
  • 10
2

The correct answer has been given, the photo does not have a dynamic range, it has only recorded the dynamic range of a scene or the specific portion of it. That being said, I will offer some unsolicited advice.

I would say that understanding dynamic range in general and specifically the dynamic range of the scene you are photographing will serve you much better than trying to do post production on a less than optimum photo. ( i have taken my share of those ) You need to know the dynamic range of the scene you are photographing AND what portion of it you want to capture, because your camera is incapable of recording the complete range of light in some/many situations.

I would suggest you study and understand if not implement the Zone system formulated by Ansel Adams and Fred Archer. You can find much info by more articulate folks then myself but i will try to give you my limited layman's understanding of it.

The Zone system will allow you to assign a reference number to a portion of the dynamic range in the scene you are photographing so that you can adjust your exposure to place that particular part of the dynamic range with the amount of exposure that fits your vision of what the final photo should look like.

Meaning if you want a certain clouds shadow falling on a mountain to be just gray enough to have details but not completely black then you can achieve that by knowing where that shadows falls within the zones of the scene and adjust your aperture and or shutter time to record it perfectly in relation to the dynamic range you can/want to record of that scene.

Another way to put is that your brain knows that your camera may not be able to record the complete dynamic range of a scene so you tell the camera what section of the dynamic range you want to capture and further more make sure that the most important part of the scene is exposed exactly where you want it within the recordable porting of the dynamic range you are recording.

All of that is not to say that post production does not serve a purpose. For Adams and others the zone system was only the first step. Knowing how to manipulate the development of the film and how print The Negative is just as important. Understanding The Camera will help you create the best photo/negative possible.

Alaska Man
  • 3,608
  • 10
  • 17