dealing with stray light from behind the subject

[forestgnome]

The earling morning light was soft and wonderful, but after a half hour the mist was lifting and some light came through the trees. The sun was about 45 degrees(?) to the right of the subject. I tried a polarizor but it didn't seem to do anything.

You can see a slight milky haze around his hip/rear leg, and his left rear knee and left front knee.

Please share tips on dealing with this.

 

[NeoAkela]

That is an amazing photo, Forestgnome... I know that doesn't answer the question, but wow! I wouldn't have noticed any blemishes if you hadn't mentioned it. What a pose!

Looks like the sun was reflecting up off the water onto the Moose's legs... not sure how you would counter that unless you could get the Moose into a studio with controlled lighting! The polarizer would fix the reflection towards the camera from the water, but not the effect the reflection had on other objects. I'd say take a few extra shots at various exposures, even though the reflection would likely still blow out those spots to some extent. You could blend parts of the darker exposures in with a bit of post-production, or use some cloning technique instead.

Of course, that is just dubious advice from my limited experience... I am sure there are others here with some real good technical solutions

 

[Mark Schaefer]

Great shot.

When there is diffuse light the sunlight is being scattered in all directions on each of the suspended water droplets in the air. The polarizer will eliminate some of the reflections on the water droplets (and on the surface of the water as NeoAkela mentions). You will in fact see some benefit with any dialing of the polarizer compared to a photo without the polarizer filter on the lens. However, it cannot eliminate all of reflections and haze. In optimal conditions the polarizer eliminates reflections and haze at 90 degree angles to your line of sight. But in diffuse light you will usually get some benefit from the polarizer, but not the maximum effect you would have in optimal conditions. FWIW, I started a topic on the benefits of a polarizer filter (in General Nackcountry): Polarizing filters, Sometimes you win, lose, or chose.

In your photo's case I do not find the bits of haze that you mention at all distracting. In fact they serve to separate the subject a bit from the background. That may not be the answer you are looking for, but I believe it is valid in your photo. You also got the slight halo effect on the edges of the moose's body, which is the same as the effect that professional photographers seek to obtain in portrait photographs using back light conditions (possibly with just a ping of fill flash).

 

[Grumpy]

This is a fine moose picture, indeed. The light and lighting are, simply, wonderful.

The rim lighting, as Mark S has observed, helps separate the subject (moose) from its surroundings. Imagine what would happen without those delightful highlights on the antlers, the subtler bright outline on the ears, that nice highlight on the hump of the moose’s shoulder and down to the rump, on the legs, along the belly and around the muzzle: the animal would blend in rather than stand out from its background, and the picture would be relatively ho-hum rather than outstanding.

The only thing that seems to be missing (on my monitor, anyway) is some small “catchlights” (or, as my spellchecker would prefer, “catch lights”) in the animal’s eyes. Those could be produced either with a bright reflector or low powered flash, but using either might also create other problems in this situation.

It also would be nice to see more detail in the moose’s chest and underbelly, but that may be asking for too much. That dark hair is a light sponge, and increasing exposure to bring out detail (or even texture) there would overexpose the antlers, muzzle, etc., and background. Flash fill might help, but it would need considerable power to be effective and might create other issuesinvolving interaction between photographer and subject.

I don’t think filtration is automatically a way to improve a photo like this. Filtration may add its own problems. The effect of a polarizer, for example, forces changes in f/stop, shutter speed or ISO, any or all of which may have some offsetting negative effects on the finished picture (loss of sharpness overall due to camera or subject movement, loss of sharpness in depth, or increased electronic image “noise”).

Also, use of a filter adds another layer of glass, which may result in further diffusion (aka, "flare").

The two things I recommend without hesitation in this (and every other) situation are (a) use of a lens shade, and (b) being meticulous about keeping the lens’ surfaces clean and free of scratches or dings. Both reduce the amount of diffused, non image-forming light reaching the electronic sensors, and help keep image contrast up. Loss of contrast is the real potential bugaboo in delightful lighting situations like this.

G.

 

[bikehikeskifish]

I'm no expert, but I certainly do not see a "problem" in this photo. In fact the separation from the background makes the moose almost seem three-dimensional. This is an outstanding photograph!

Tim
p.s. Is that "Big Daddy"?

 

[Tim Seaver]

Great shot - his expression and the slight tilt of his head say "You want a piece of this?" I can't say I even notice any problem with the backlit "furry edges".

 

[w7xman]

Backlighting of a furry or feathery subject IMO is rather idea, as the translucency of the edges creates a nice separation between the subject and the background.

The problem though is that when you have a bright background, and very dark moose, you have a contrast problem that can (doesn't in this case though) exceed the normal dynamic range of an image. Unfortunately there is nothing that you can do in the field to remedy this...

You did right to expose for the moose here, and he looks great. The problem is that the background is bright, flat, and lacking contrast. I did some quick editing, using three layers, and quick masks.

Layer 1: In layers, take out some highlights, and slide midtones and shadows to the right. Then, while in that layer using a black brush, paint the mask over the moose at 100% opacity to remove the effect from the moose, leaving a more contrasty background.

Layer 2: Reduce saturation using hue/saturation. Again, mask out the moose leaving him fully saturated.

Layer 3: Using the dodge and burn technique I showed you earlier in the year, open a new layer with a midtone, and 'paint' a burned in vignette and tone down other bright spots on the background.

Note...I did nothing to the moose at all...only adjusted the background to yeild a more pleasing scene that puts the focus on the moose without distractions.

Thoughts, questions/comments and disagreements welcome!

 

[Grumpy]

My monitor reveals (to me) very little difference between the original provided by forestgnome and the reworked version provided by w7xman.

To me, the original handled the relatively great dynamic range (of brightness values) in the scene very well, considering that the capability of the camera sensor in that respect is limited.

This is an appropriate time to note that the lighting situation in the moose picture is very similar to what w7xman found in the recent and wonderful lupine meadows photos. Both remind us that photography calls for being conscious of the quality of light,

G.

 

[bikehikeskifish]

My monitor reveals (to me) very little difference between the original provided by forestgnome and the reworked version provided by w7xman.

Whew. I'm glad I'm not alone in that opinion

Tim

 

[The Unstrung Harp]

I keep looking at this photo. I love the colors and the...intimacy, perhaps? It is a great shot, and I like the natural "halo" effect.

I think the only thing I would change (and I'd think it'd be an easy fix, if you wanted to do it) is to add some halo-effect to the moose's right eye. Or some sort of contrast. The left eye looks good and alive, but the right could be brought out a little more.

Disclaimer: I don't necessarily know what I'm talking about.

 

[w7xman]

Whew. I'm glad I'm not alone in that opinion

Tim

Yeah...it's pretty subtle...

The edit has a slightly darker, more contrasty background, and a vignette focusing the eye on the moose a bit more...

Side by side:
First the orig, then the edit:

 

[forestgnome]

Thanks very much for all the great comments and suggestions. Jim, thanks for the editing work and comments. It's good practice.

Yes, that's Big Daddy. I really adore this bull. He's probably about 12 years old. It as a wonderful early morning scene. You could hear him breathe.

I'm happy that I was able to shoot this at 400ISO and still have a fast shot at1/100. Lens was at 44mm, f/4.5.

Thanks for all the help.

 

[hikingmaineac]

You can see a slight milky haze around his hip/rear leg, and his left rear knee and left front knee.

Actually according to a game of Trivial Pursuit I played last night... elephants are the only mammal with 4 knees.

... besides that - awesome photo!

 

[ALGonquin Bob]

Me thinks you're too critical of your image. Me likey moose picture.

Seriously, the original is very nice. In this age of digital manipulation, I think too much emphasis is put on what you can do to the picture after the capture. Composition, lighting, and exposure are the main concerns, and with RAW capture, exposure isn't as important as it used to be. It's an excellent image.

 

[Mark Schaefer]

You can see a slight milky haze around his hip/rear leg, and his left rear knee and left front knee.

Actually according to a game of Trivial Pursuit I played last night... elephants are the only mammal with 4 knees.

Yes, moose are ungulates, using unguligrade locomotion, walking on their hooves. That is similar to digitigrade locomotion, walking on toes. Those backward knees are really their ankles. I believe that most unguligrade and digitigrade mammals technically have knee /elbow equivalents as part of their hindquarters / forequarters. Humans and elephants are among the rare mammals that walk on the soles of their feet using plantigrade locomotion. Not that any of this matters much to man nor beast. And we all knew what forestgnome was referring to.

Edit: While checking my facts I see that elephants are actually in a special category of semi-digitigrade and semi-plantigrade locomotion. I think I will punt rather than fumble any more of an explanation.

 

[ALGonquin Bob]

The only thing that seems to be missing (on my monitor, anyway) is some small “catchlights” (or, as my spellchecker would prefer, “catch lights”) in the animal’s eyes. Those could be produced either with a bright reflector or low powered flash, but using either might also create other problems in this situation.

I respectfully disagree, Grumpy. Adding catch lights, something you'd expect in a human portrait, is inappropriate in animal "portraits", and I think that it's absurd to add that to an animal photograph.

 

[Grumpy]

Me thinks you're too critical of your image. Me likey moose picture.

Seriously, the original is very nice. In this age of digital manipulation, I think too much emphasis is put on what you can do to the picture after the capture. Composition, lighting, and exposure are the main concerns, and with RAW capture, exposure isn't as important as it used to be. It's an excellent image.

I'll agree with most of what ALGonquin Bob says here, but we part company on the matter of exposure being less important with RAW capture than it once was. Yes, it is possible to "salvage" OK images from mis-exposed RAW "captures," but only within limits that I find similar to those of the film days. The same problems crop up with both underexposure and overexposure.

The best we can do is hit the exposure nail on its head. I find myself more and more often overriding the automatic exposure features on my DSLR and using manual settings to get it right, especially under high dynamic range and odd light color conditions. The automated exposure systems are very good, but far from perfect.

G.

 

[ALGonquin Bob]

A RAW image file should have the equivalent of 40 stops of exposure latitude - that's one of the big advantages of capturing in that mode. J-peg, however, has limited latitude, similar to slide film.

 

[Grumpy]

I respectfully disagree, Grumpy. Adding catch lights, something you'd expect in a human portrait, is inappropriate in animal "portraits", and I think that it's absurd to add that to an animal photograph.

I do not propose adding catchlights to subjects' eyes artificially (via processing tricks), either with human or animal subjects. But I do observe that catchlights created by the lighting in a scene do "animate" both human and animal subjects -- makes them look more alive. As a hunter, I have had critters die in my hands, at my hands, and will testify that one literally sees the light go out of a creatures' eyes at the moment of death.

G.

 

[DougPaul]

A RAW image file should have the equivalent of 40 stops of exposure latitude - that's one of the big advantages of capturing in that mode. J-peg, however, has limited latitude, similar to slide film.

RAW files are linearly sampled which limits the maximum dynamic range to approximately one stop per bit. The largest number of bits that I am aware of in DSLRs is 14 bits, thus 14 stops. (The RAW file typically has the same number of bits per sample as the A/D converter.)

The sensor, optical, and electronics systems can be far more limiting: some measurements of several recent high-end digital cameras give a highest dynamic range of 11.7 stops. The same author gives the greatest theoretical range of a sensor (5D Mk II) of ~14.7 stops.
http://www.clarkvision.com/imagedetail/digital.sensor.performance.summary/#dynamic_range
http://www.clarkvision.com/imagedetail/index.html#sensor_analysis

JPEG uses an 8-bit per sample representation, so it is limited to 8 stops. 8 stops post processing (eg for printing or screen display) is fine, but can be inadequate before processing.

BTW, dynamic range in film and sensors is often (generally?) denoted in D (log base 10) rather than stops (log base 2). log10=.301*log2, so 11.7 stops is equal to a D of 3.5.


The human eye has a greater effective range because we, unlike a camera, can change the "exposure" as we look at different parts of a scene. (One could argue that this ability to adapt to different parts of a scene is similar to what one can do in post-processing of an image.)

Doug