Photography on moving ships?

jerry12953

Suspended / Banned
Messages
12,421
Name
Jeremy Moore
Edit My Images
No
I'm thinking of going on one of the Hurtigruten Northern Lights cruises next year - these - and i wondered if anyone has any experience of taking long exposures on moving ships. I know it wouldn't be ideal but as the Northern Lights are moving themselves would a bit of camera movement really degrade the image that much? And what about engine vibration?

Thanks in advance.
 
Not sure if its Nikon only, but the vr (vibration reduction) on our vr lens has a three way mode to be used when the base you are on is moving can even be used on a tripod i think you have to tell the camera your using a tripod. also wide angle lenses would not show as much movement as tele hth mike.
 
Some good points there...makes me think it might work. Maybe I should contact Canon about their IS (VR).

Thanks
if you have the boxes/instructions that should tell you not sure which Canon lenses you have?
 
been searching online but can,t find info on mk1 I.S. tried youtube for vids but nothing there.doh!
 
what i have seen about the 24-105 is that it was a pretty good lens for the money but no one talking about the I.S.
 
Such a ship will be moving at anything from 10 to 30 miles per hour.
This is exactly the same as the subject moving at that speed.
How much it moves will depend on the direction of travel relative to the field of view, And the distance to the subject.
Inevitably this will blur the image on a time exposure. It will not be isignificant.
 
I'd be using the 24-105 Mk1 IS. I also have the 17-40 but it doesn't have IS.
You don't want to use the 24-105 for Northern Lights. It's not wide enough. The wide end of the 17-40 would be OK for focal length, but f/4 isn't ideal and of course you don't have IS.
 
You don't want to use the 24-105 for Northern Lights. It's not wide enough. The wide end of the 17-40 would be OK for focal length, but f/4 isn't ideal and of course you don't have IS.


Good point, although it depends how far away they are. I suppose they could be right overhead....

I've not booked yet but I may need to rent a lens......;)
 
Might be worth trying to stack multiple shorter frames rather than one long exposure?
I saw a YouTube video on using that effect for faking long exposure without a big stopper or similar.
The video showed Photoshop autoaligning layers which could compensate for the boat movement. I had a play in GIMP doing it manually with a fountain but it was quite a faff.
I suppose it depends how long an exposure you’d be trying to recreate/mimick.
 
Might be worth trying to stack multiple shorter frames rather than one long exposure?
I saw a YouTube video on using that effect for faking long exposure without a big stopper or similar.
The video showed Photoshop autoaligning layers which could compensate for the boat movement. I had a play in GIMP doing it manually with a fountain but it was quite a faff.
I suppose it depends how long an exposure you’d be trying to recreate/mimick.


Wow! I'll have to give that some thought......
 
The 24-105 f4 IS Mk1 does not have tripod sensing, you need to turn IS Off when its on a tripod-I know from bitter experience.
 
For what it is worth a few years ago I took a tripod on-board a ship (I did ask for permission before hand) with the idea of taking some HD video footage from the stern of the ship.

On returning home much of the footage was of no use because of vibration even though the tripod appeared to have sturdy rubber feet.
 
You may not agree with what the chap is saying about needing filters but he explains the stacking alternative well.
This is the video jump to 2m40s to see the technique.

View: https://www.youtube.com/watch?v=YcZkCnPs45s


My opinion of Tony Northrup has just gone down a couple of notches after watching that........;)

However I AM interested in the technique you/he describes for stacking images rather than using long exposures. I don't use photoshop so that is a disadvantage but can anyone else understand and/or explain if the technique would translate into the situation I might find myself in? I can't get my head around it.

ie photographing the Northern Lights from a moving ship with possible/probable engine vibration.

Otherwise it looks like I'm looking for 20mm, f1.8 lens with 4-stop image stabilisation, and I can't find one........

Thanks VERY much for the suggestion, A!
 
I AM interested in the technique you/he describes for stacking images rather than using long exposures. I don't use photoshop so that is a disadvantage but can anyone else understand and/or explain if the technique would translate into the situation I might find myself in? I can't get my head around it.

ie photographing the Northern Lights from a moving ship with possible/probable engine vibration.
That's fascinating. I've always considered ND filters to be a bit of a faff, and I never like lugging tripods around at the best of times. So I really like the idea of synthesising a long exposure as demonstrated here. I'm definitely going to give it a try.

Would it work for the Northern Lights? I don't see why not. At the end of the day all you're doing is capturing and counting photons, and I don't think it necessarily makes a difference where those photons come from.

Could you do it on board a ship? Yes, I should think so, so long as you can align the images properly. Photoshop can clearly do that; I don;t know whether Gimp can, but if it can then that's a free solution. One potentially critical issue though is that on a ship you would probably get more camera movement from one image to the next than you would on dry land, so there would probably be more "wasted" space around the edges where the images don't align perfectly. That suggests using a wider lens than you otherwise would, if you want to make this technique work.
 
I'm thinking of getting hold of Affinity Photo in the near future which I gather is a great Photoshop alternative at a very reasonable price; I imagine it would have an equivalent feature for stacking images. Looks like there may be a avenue to explore there.

Let us know how your experiment goes Stewart.
 
FWIW you can do the same trick manually in GIMP - it's a faff.

I had one test go here, camera propped on a post...
https://www.avforums.com/threads/nd-filters-advice-please.2089728/#post-25027106
I didn't persevere with the image as it wasn't interesting, just a test.
I realise now I missed his suggestion that you use a slower shutter speed around 1/5th or 1/6th might make it impractical.

I tried it again on holiday but hand held this time, this introduced not just some linear movement but also so rotation which I was too lazy to adjust for so I dropped those frames.
Again the shutter speed is too high.
https://www.avforums.com/threads/mi...-images-welcome.1633919/page-82#post-25452576

Aligning the images is awkward and then you need to calculate the necessary transparency settings for each layer manually which GIMP doesn't actually seem to like very much and applies some rounding.
If I were seriously interested I'd want a solution that at least auto aligned the layers.

(I hope there's no problem with linking to my images on another forum)
 
I suppose what I'm wondering is whether a (relatively) static object photographed by a moving camera would be equivalent to a moving subject photographed by a static camera; in terms of image stacking, that is........

Maybe its time to start a new thread......we've moved quite a long way from my original one......:)
 
I suppose what I'm wondering is whether a (relatively) static object photographed by a moving camera would be equivalent to a moving subject photographed by a static camera; in terms of image stacking, that is........

Maybe its time to start a new thread......we've moved quite a long way from my original one......:)
Are we off topic really?
I would imagine you will have some fixed points of reference the horizon and stars that you could use for alignment.
Then you'll have the Northern lights themselves which will be a moving subject but in roughly the same area relative to the horizon..
Finally you'll have some stuff moving relative to the horizon, anything on the ship including the camera.
If the shutter speed is fast enough to deal with the motion of the ship so each individual frame is sharp (enough) then I can see it working but you might spend your exciting Northern Lights moment mucking around with your camera rather than enjoying the moment.

I can't think of a sensible way to test the process though other than trying to do some plane trails or satellite passes while deliberately moving the camera ?!? :confused:
 
Is there any reason you can't use video and take a still from that?

Having read through I think there may be a few things to consider. With regards to movement and using a long exposure, I'm not sure if this would work as their will be too many moving elements, you will have the ship moving no o my forwards but also up and down. There is also the vibration from the engines coming through the decks.

With regards to the image stacking I'm not sure how this would work, because surely part of the reason for the long shutter is due to the lack of light. Doesn't matter how many images you have if they are all dark to start with.
 
I would think the video idea may be sound. One could perhaps extract a number of stills from the footage and then blend them together. I have been on several cruises and long ferry trips and never used a tripod whilst the ship is underway - always hand held.

It might also be worth contacting Hurtigruten to see if they have any advice.
 
By long exposure do you mean just long enough to capture the lights or really long exposure? I took the pic below with a X100S so 35mm equiv on a ship at F2 3.5 seconds ISO 6400 hand held braced against a large pillar. It was in transit on a very smooth sea. All ships vibrate to some extent, some more so than others which can effect the tripod, also ship motion can b****r it up. Whilst the longer exposures won't really bother the moving aurora the bigger issue is any land, particularly if there is distant settlements lit up and in particular the stars which will easily show movement.

 
That's the sort of thing I'd hope to capture. What percentage of keepers would you say you got using this technique?

Did your lens/body have IS?

The keeper rate was pretty low. But I had no IS and that was the max ISO the body does so I was quite limited. If you click on the pic to go to Flickr I have 2 other shots which were around 1.5 - 2 secs but which aren't quite so sharp. It was quite hit and miss. So with IS and being able to push the ISO I'd expect you to get a better keep rate. Being wider at 24mm would help too.
 
FWIW you can do the same trick manually in GIMP - it's a faff.

I had one test go here, camera propped on a post...
https://www.avforums.com/threads/nd-filters-advice-please.2089728/#post-25027106
I didn't persevere with the image as it wasn't interesting, just a test.
I realise now I missed his suggestion that you use a slower shutter speed around 1/5th or 1/6th might make it impractical.

I tried it again on holiday but hand held this time, this introduced not just some linear movement but also so rotation which I was too lazy to adjust for so I dropped those frames.
Again the shutter speed is too high.
https://www.avforums.com/threads/mi...-images-welcome.1633919/page-82#post-25452576

Aligning the images is awkward and then you need to calculate the necessary transparency settings for each layer manually which GIMP doesn't actually seem to like very much and applies some rounding.
If I were seriously interested I'd want a solution that at least auto aligned the layers.

(I hope there's no problem with linking to my images on another forum)


i've queried this elsewhere and been told that it wouldn't work. The reason being that one would be taking a series of VERY underexposed images and averaging out the exposure to get a finished image which would still be VERY underexposed.

But what about a simple multiple exposure technique where the process is additive?
 
i've queried this elsewhere and been told that it wouldn't work. The reason being that one would be taking a series of VERY underexposed images and averaging out the exposure to get a finished image which would still be VERY underexposed.

But what about a simple multiple exposure technique where the process is additive?
You've been advised correctly - thinking a bit harder the technique I've suggested is trying to cope with too much light for a long exposure, not too little.

It might be possible to stack images and use an additive mode for each layer. I've read that lots of asto photographs are stacked form multiple images but that's the sum total of what I know :)

It would be possible to test - set up a night sky/star shot on a tripod and find a sensible length of exposure to get a bright image.
Then take a series of shorter exposures that add up to the same exposure - for illustration One 1 second exposure = Ten 1/10th exposure frames.
Import them into GIMP as layers and change the mode - I'm guessing "Lighten only".
My knowledge of astrophotography is next to nothing - if you need a 3.5 second ISO 6400 exposure (as the image above) then you're going to need a lot of images to freeze any ship motion induced blur in the stars and still get a cumulative 3.5 seconds
If 1/10th was fast enough you'd need 35 images.
I know from my limited stacking experience in GIMP manually aligning 35 layers for rotation and translation is going to take ages and drive you nuts if the layers themselves are heavily under exposed so there's not much to use as a marker (e.g. a bright star or lit object on land).

I'm way out of my own experience here so probably just making things worse, sorry.

You probably need some very effective stabilisation rather than attempting this.
 
You've been advised correctly - thinking a bit harder the technique I've suggested is trying to cope with too much light for a long exposure, not too little.

It might be possible to stack images and use an additive mode for each layer. I've read that lots of asto photographs are stacked form multiple images but that's the sum total of what I know :)

It would be possible to test - set up a night sky/star shot on a tripod and find a sensible length of exposure to get a bright image.
Then take a series of shorter exposures that add up to the same exposure - for illustration One 1 second exposure = Ten 1/10th exposure frames.
Import them into GIMP as layers and change the mode - I'm guessing "Lighten only".
My knowledge of astrophotography is next to nothing - if you need a 3.5 second ISO 6400 exposure (as the image above) then you're going to need a lot of images to freeze any ship motion induced blur in the stars and still get a cumulative 3.5 seconds
If 1/10th was fast enough you'd need 35 images.
I know from my limited stacking experience in GIMP manually aligning 35 layers for rotation and translation is going to take ages and drive you nuts if the layers themselves are heavily under exposed so there's not much to use as a marker (e.g. a bright star or lit object on land).

I'm way out of my own experience here so probably just making things worse, sorry.

You probably need some very effective stabilisation rather than attempting this.


Thanks again. Despite the advice I've been getting I'm sure it is possible to do what you originally suggested, only to use some kind of additive method rather than the averaging method, the technique for which is widely available on the internet.

As you say, it seems to relate particularly to astrophotography, but I'm really struggling to find a simple explanation of the technique in non-scientific terms. I think most astrophotographers are scientists and they assume us normal folks understand all the jargon! Anyway, I shall keep searching.

I think I will post an enquiry in the post-processing forum, if only to get some attention from processing experts who may not have seen this.:)
 
My advice would be to try a few 3 or so second exposures braced against something solid and see if you get any keepers. Then try a minute or so of video and see if your kit has done the AB any justice. Then sit back and let your Mk 1 eyeballs capture the show in all its glory! That's IF they make an appearance during your trip. (Mum and Dad did it a few years back and didn't see them.)

Good luck!
 
My advice would be to try a few 3 or so second exposures braced against something solid and see if you get any keepers. Then try a minute or so of video and see if your kit has done the AB any justice. Then sit back and let your Mk 1 eyeballs capture the show in all its glory! That's IF they make an appearance during your trip. (Mum and Dad did it a few years back and didn't see them.)

Good luck!


If there is no sighting they now offer a "free trip" guarantee; but I imagine there will still be many hours of darness with full cloud cover and enforced idleness.

I haven't booked yet but there is still availability for a single room on one trip at the end of Feb.
 
Back
Top