I wrote a simple shell script[0] to scrape and output the latest image.
It uses the tiles from their online satellite map[1], and can output images in increments of 1x1, 2x2, 4x4, 16x16 tiles (each tile being 550px by 550px). Here is an example with 2x2 [2]
If you have any suggestions or bugfixes, feel free to fork or comment.
EDIT: Also works for a single tile[3], also clarity.
I wish there were a pair of these satellites separated by a couple thousand miles/km, so we could view these images with next-gen, hi-rez VR goggles and see the surface and cloud topography with enhanced depth.
Earth's diameter is 12742 km. The clouds are about 0.15 % of this size, relatively speaking. The full images are about 11k by 11k, so we can say that 1 px is roughly 1 km. This means that the clouds will be on the scale of 20 pixels, roughly.
Just give it a shot, you could translate it line for line into Ruby, and use ImageMagick or something. Don't sell yourself short, it won't take as long as you think!
Here is a 11000x11000 = 121 million pixel 'full disk' image [1] (lower resolution at [2]) captured at 0340 UTC on 7 July 2015 by the Himawari-8 satellite, which I found on a blog post [3]. Real-time images from the weather satellite can be found here [4].
Nice. Japan needs better weather data; too many hurricanes and too much coastal development. From geostationary orbit, the resolution has to be low, but it's always on.
The US has two geostationary weather satellites, which are usually parked roughly over Panama and Hawaii. Neither has good coverage of Japan. Korea's COMS satellite does, though. China has several, including one that's usually pointed roughly at Taiwan.[1] Right now, you can see the hurricane that's due east of Shanghai.
I didn't realise they were lacking in weather data. Seeing tangentially related projects like their supercomputer Earth Simulator Project [1] led me to (falsely it seems) believe they were likely also sufficient in that area.
Beginners question - does their inclination (prohibitively) affect the efficiency of lower orbit weather sats?
edit: The largest image resolution seems to be 800x800 (well, 800x815, with the attribution footer). I wonder if the new satellite's imagery isn't part of this feed yet? I hope that's the case, because it would be incredibly cool to be get such high-res images.
I wrote a script which downloads the latest image in a large resolution[0]
If you can set your wallpaper from the commandline, or have the wallpaper refresh from an image on disk, you could use it for a live wallpaper.
I'm also going to have the scraper check if the image has changed every 10 minutes and have it update at [1] if it has, but the image will only be in 1,100px x 1,100px resolution.
Privacy is on a sliding scale. There are certain assumptions we rely on when we go out in public.
Imagine, for example, some future where you were effectively under constant surveillance when in "public". Where your every action, every movement, every utterance was recorded, categorized, transcribed, data mined, and then put into a feed that anyone could subscribe to. That's what the complete absence of public privacy looks like, and it's not great.
But, AIUI, those cameras aren't networked together. They're all owned by different people. TV shows like Torchwood that show some agency tracking someone across the city using CCTV are showing something that can't possibly happen in real life, because there is no one centralized authority (or even handful of authorities) that have access to most of those cameras. Not to mention the cameras themselves are just video, and not particularly stellar quality at that.
Of course, I could be wrong, maybe someone in London really can track someone across town using the CCTV cameras, but the above is what I understand from what other people have explained in the past.
It won't because these are geostationary satellites (if I read the post correctly). So you'd need at least 3 of these to get a good image and that's not even considering some of the bigger issues with this. I also don't think the resolution is on par. But the images will be really cool to see.
Yup that's what I was getting at, my first thought was that they'd be competing then I looked at the use cases and I wanted to point out that they were different, but I might have been the only one who thought that in the first place :-)
Are there any other sources with constant up to date imagery like this? This looks great and would like to see other current satellite images, and maybe use them for visualizations or some projects.
Satellite TV and communications satellites commonly do this -- otherwise your satellite dish would have to track the satellite across the sky as it moves.
Defense Support Program satellites are also in stationary orbits so they can watch fixed regions for nuclear missile launches with infrared cameras.
The resolution of the "full disk" (i.e. whole earth) natural color images appears to be 11000x11000 pixels every 10 minutes. I can't find any realtime access to these images though - could anyone else?
They do have a cloud service for disseminating the imagery, but only for "official use":
"Until Himawari-8 becomes operational, NMHSs wishing to release Himawari-8 data and products to the public are requested to consult with JMA beforehand."
Edit: Here is at least a tile-zoomer with some sort of realtime access to high-res imagery: http://himawari8.nict.go.jp/
These are great! I've been looking for something like this for a long time and I could never find anything out of NOAA that had this kind of time resolution. Here about a big storm on the east coast? What does it look like from space? Gotta wait to see whether someone will do a press release.
You might find http://realearth.ssec.wisc.edu/ useful. Just be sure to click on either the "Presets" or "All" tabs within the "Layers" section.
Oh, and if you decide to use the animation feature just try to be patient with the way it loads in the individual times--once you've pulled down everything the animation should look a good bit nicer.
It uses the tiles from their online satellite map[1], and can output images in increments of 1x1, 2x2, 4x4, 16x16 tiles (each tile being 550px by 550px). Here is an example with 2x2 [2]
If you have any suggestions or bugfixes, feel free to fork or comment.
EDIT: Also works for a single tile[3], also clarity.
[0] https://gist.github.com/Syrup-tan/1833ba1671c7017f0d59
[1] http://himawari8.nict.go.jp/
[2] https://denpa.moe/~syrup/himawari8.png
[3] https://denpa.moe/~syrup/himawari8-single.png