No exactly super sharp. And drop the nonsense about the video not being able to show it. The fact that it doesn't even look sharp in the video, show how weak the result is. I tried out a lot of superresolution programs, and it's over all quite disapointing.
480X480 isn't sharpest, but from 48X48 - aka super pixelated puny image? 100 times improvement in resolution is very nice, try finding one like that (and wanna post link to it for varification?) If not, STFU.
@migleycow - hey, it's not exactly sharp. I could interpolate the pixels to make the moon the size of a wall while loosing the pixelated look, but that doesn't mean it would be a million times improved or sharper. I do admit that it got sharper, but not sharp enough for the picture's size. So you can't just say it's "100 times improved" from the pixel difference. 5 times improved is a more realistic number. It's possible to get a (significantly) sharper image, but you need many more pictures.
Using this technique they went from a 50x50 sized picture to a 500x500 :P The video cant really show you the improvement.. check the website perhaps it shows more.
I don't use photoshop, but I'm pretty sure that using antialias/sharpen does not combine 60 images into one. Perhaps you thought that they only took one low resolution image of the moon.
This technique has been around for a while. Do Google searches for the terms "astrophotography" and "webcam" and you'll see that plenty of amateur astronomers use this for planetary photography in particular, using modified webcams and readily available astronomy software.
My only worry is that this is software modified imagery and as such subject to errors that might be accepted as correct in research using similarly modified imagery and slowly distort calculations based on deduced information that was inaccurate, however minor (in a similar way to errors in a floating point chips calculations might be miniscule but have a much greater effect over deductions made using these figures).
since images are to be observed by human eye the small inaccuracy wont matter but if you are planning to land in the moon an SLR camera would not be a good choice to pick a landing zone
This is kinda like when they take a bunch of photos of average looking people, combine them, and the result is 'someone' who looks beautiful ('someone' in quotes because that person does not exist). Basically, flaws get cancelled out when averaging.
Good question, it does indeed rotate on it's own axis. In fact it rotates once every 27 days 7 hours and 43 minutes. The same time it takes to orbit the earth. The net effect is that as it revolves around the earth it's own rotation is such that the same side is always presented to the earth.
It isn't a coincidence, it's called "tidal locking". Basically the moon isn't perfectly spherical, and a bulge is attracted toward the Earth by their mutual gravity. The bulge always faces the Earth, and therefore we always see the bulge side.
Very nice. However, it would also require a technology that took 60 pictures in an instant. It might actually be cheaper to just implement higher resolution. However, I can see the advantages of this algorithm for various specific situations.
... I thought that was obvious, I mean, I've been using Photoshop to combine blurry photos I've taken often to get slightly better quality. Nothing special.
THAT WASN'T EVEN SCIENCE IT WAS PHOTOSHOPPE!!!
RoboticusMusic 2 months ago
maybe you can use this technology to upscale your 240p video to 1080p ¬¬
liquidus2172 11 months ago
1:0 to CSI fanboys
xzacq 1 year ago
i wish i was a werewolf
scarface1567 2 years ago 3
WTFHAX!!?!!?
Ozzy0sbourne 3 years ago
and the ring is a consequence of what exactly?! or what is that?
this shouldn't be viewed as an image, but rather some 3d or 4d thingy...
sidthemyth 3 years ago
AWE inspiring!
I love it and the idea is so simple.
Is there software I can get for this?
Thanx so much for posting.
akaalgorithm 3 years ago
To get that improvement they had to have 60 pictures of the moon all in slightly different positions. Not exactly practice.
artifactingreality 3 years ago
0:56 is a huge improvement in resolution. This would be very useful for hardware limitations.
FoodCubes 3 years ago
No exactly super sharp. And drop the nonsense about the video not being able to show it. The fact that it doesn't even look sharp in the video, show how weak the result is. I tried out a lot of superresolution programs, and it's over all quite disapointing.
jelleoelle 3 years ago
480X480 isn't sharpest, but from 48X48 - aka super pixelated puny image? 100 times improvement in resolution is very nice, try finding one like that (and wanna post link to it for varification?) If not, STFU.
migleycow 3 years ago 3
@migleycow - hey, it's not exactly sharp. I could interpolate the pixels to make the moon the size of a wall while loosing the pixelated look, but that doesn't mean it would be a million times improved or sharper. I do admit that it got sharper, but not sharp enough for the picture's size. So you can't just say it's "100 times improved" from the pixel difference. 5 times improved is a more realistic number. It's possible to get a (significantly) sharper image, but you need many more pictures.
danielodors 2 years ago
Not that well resolved...
cyborgtroy 4 years ago
Using this technique they went from a 50x50 sized picture to a 500x500 :P The video cant really show you the improvement.. check the website perhaps it shows more.
mamolian 3 years ago 4
This comment has received too many negative votes show
So someone figured out how to operate the antialias/sharpen filters in photoshop...
Ichinin 4 years ago
I don't use photoshop, but I'm pretty sure that using antialias/sharpen does not combine 60 images into one. Perhaps you thought that they only took one low resolution image of the moon.
newpip14 4 years ago 4
not only do you not understand how cameras work, but also fail to grasp photoshop as well. congratulations, you are really stupid.
buttface1202 4 years ago
This technique has been around for a while. Do Google searches for the terms "astrophotography" and "webcam" and you'll see that plenty of amateur astronomers use this for planetary photography in particular, using modified webcams and readily available astronomy software.
TravisMorien 4 years ago 6
What?! Who gave that a thumbs down, he's completely right.
terser 3 years ago 2
My only worry is that this is software modified imagery and as such subject to errors that might be accepted as correct in research using similarly modified imagery and slowly distort calculations based on deduced information that was inaccurate, however minor (in a similar way to errors in a floating point chips calculations might be miniscule but have a much greater effect over deductions made using these figures).
j1mb0jones 4 years ago
I think the information its not deducted but its an average of all 60 images
alejo0823 4 years ago
My worry's still there as there's still an inherent inaccuracy however small.
j1mb0jones 4 years ago
since images are to be observed by human eye the small inaccuracy wont matter but if you are planning to land in the moon an SLR camera would not be a good choice to pick a landing zone
alejo0823 4 years ago 2
This is kinda like when they take a bunch of photos of average looking people, combine them, and the result is 'someone' who looks beautiful ('someone' in quotes because that person does not exist). Basically, flaws get cancelled out when averaging.
badbuddhist 4 years ago
Wrong... averaging only removes noise, but it doesn't remove blur and doesn't increase resolution (48px->480px)
BamBoom01 4 years ago 2
... to me that super-resolution stuff looks different to simple averaging (in response to badbuddhist and alejo0823)
BamBoom01 4 years ago 2
so let me be a dumby, the moon does not rotate?? on its own axis? as the image would indicate?
archaedemos 4 years ago 2
Good question, it does indeed rotate on it's own axis. In fact it rotates once every 27 days 7 hours and 43 minutes. The same time it takes to orbit the earth. The net effect is that as it revolves around the earth it's own rotation is such that the same side is always presented to the earth.
LukeSkyscraper 4 years ago 3
i gotta say that is So Damn Coincidental, that it blows my mind about the universe.
archaedemos 4 years ago
It isn't a coincidence, it's called "tidal locking". Basically the moon isn't perfectly spherical, and a bulge is attracted toward the Earth by their mutual gravity. The bulge always faces the Earth, and therefore we always see the bulge side.
TravisMorien 4 years ago 4
ok so the Moon has a Bulge... hmm i guess that explains it.. but i still gotta wonder how the hell it got here/locked in like that (:
archaedemos 4 years ago
Very nice. However, it would also require a technology that took 60 pictures in an instant. It might actually be cheaper to just implement higher resolution. However, I can see the advantages of this algorithm for various specific situations.
massenergy 4 years ago
To echo previous comments, the resolution hasn't improved all that much. Would be a hand thing to employ on our badly taken Christmas snaps though!
PowerMetalDom 4 years ago
It might also be useful in obtaining better images from security cameras
Jmcenanly 4 years ago
Not true, because you'll have to get numerous pictures of each moment. Security cameras don't shoot so many frames so fast.
AwkwardSquad 4 years ago
wtf. no such thing as atomic bombs? so those videos or bombs going off are CG or sumthin? what did america drop on Nagasaki?
ladyboybassist 4 years ago
idiot
britoca 4 years ago
... I thought that was obvious, I mean, I've been using Photoshop to combine blurry photos I've taken often to get slightly better quality. Nothing special.
trav10852 4 years ago
But you say "slightly better quality". They improved resolution by a factor of 10x in both dimensions.
yubeaut 4 years ago 2
culo
charliepoli 4 years ago
cool
lisalovesthekillers 4 years ago
at 00:53 it looks like theres a face with sunglasses haha
magicbitch1995 4 years ago