Slashdot Mirror
Development of OS Satellite Image Processing/Mapping
Ken Melero writes "ImageLinks, in collaboration with Federal Agencies and Florida Tech is developing Open Source software for satellite image processing and mapping. The Open Source Software Image Map (OSSIM) experiment will be hosted on RemoteSensing.org and will demonstrate the application of open source software development techniques in meeting government requirements. See the press release for more information. " Looks cool - GPL as the license.
Will it include instructions to connect to satelites and get pictures?
www.spaceimaging.com (the first ones to sell commercial high-res imagery, very cool site with sat photo dowonloads)
A report by the Carnegie Endownment For Internatinal Peace
An abstract is posted online with the full report available for download.on the effects of commercial High-res.
Secrets for Sale
tcd004 Here's my Microsoft Parody, where's yours.
In several years of working in physics/programming with satellite tracking data, I found out just how bad the documentation and standards were for the various formats for data on and from satelites. The images were often fine, being compressed versions of standard formats, but what we'll be looking at here, it seems, would require additionally the exact positional data for the satellite body. Here's hoping the open source approach leads to better standards and documentation of data formats for time, position relative to the earth (and orientation), and so forth.
From experience, it will more likely result in filters of resolutions with a tremendous range, both in granularity and depth, being interpolated and patched together. I'm not sure I believe that OS will get around the traditional bailing wire and duct tape aproach of engineers in the field...
-- Still waiting for the Nike endorsement
That said, I have to express a mixture of excitement and concern over what the success of this project could mean. On the one hand, it could mean that corporate culture will begin (slowly) to adapt to the Open Source model. On the other, it could mean that corporations will glean the hard work of Open Source developers and cash in, doing very little actual work themselves.
In the first case, I am happy on two counts: changing corporate culture is difficult, and if the Open Source Movement succeeds, it will be a big "feather" in our collective hat; and, corporate adaptation means that more companies are likely to open hardware specs for Linux support, and software specs for cross-platform ports.
In the second case, I have mixed feelings: having a corporate infrastructure to market and distribute Open Source works has the makings of a rather sound business model -- however, the less ethical companies out there (which, IMHO are a vast majority) may abuse the model so thoroughly that Open Source developers are alienated.
However, I cannot decide whether Open Source would be hurt, or if developers would just move thier resources to the more ethical companies. The concern is that even if the latter, the "evil" orgs can still get the work -- possibly with even less effort.
Anyone see a way around that mess?
--
We may not imagine how our lives could be more frustrating and complex—but Congress can. – Cullen Hightower
If we all could have our own satellites..
-motardo
It is a nice project and it does have application in the corporate world. Yes, I agree, it is a good. But . . .
what the open source world needs to push it over the "edge" is a project that everyone would want to have, a killer app if you would.
if, say, the open source world got together and built an openDoc office suite, then that would be worth talking about . . . 'til then, well, we'll see.
Later . . . . . . WebBug
...was impossible to trace by following the links. Assuming that the US government is involved, it is wrong.
Think about it. Tax dollars come from the people and from corporations. That includes MS employees and other software companies. They are using tax dollars to fund software written under a license that effectively tends to nationalize an industry, without any consent from the people.
This is just symptomatic of the whole leftist Clinton/Gore/Judge Jackson mentality that pervades this administration.
Once again, I have no problem if private organizations want to collectivize themselves, but when my government gets involved, warning bells start to go off. Karma to burn lately, so mod all you want.
For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
The Image Processing Workbench (IPW) of UCSB has been open source and freely available for years, and recent ports include Linux. Very powerful collection of command line Unix utilities written in C that can be pipelined together. It's specifically designed to work with remote sensing data although it does not incorporate image projection (mapping) and navigation functions. There is still a need for freely available OS tools for mapping and navigation.
I am quite civilized, and I should be brought a beer immediately. -- Bruce Sterling
But is this "satellite-related data" which requires an ITAR license for export?
Great, yes! I forgot about GRASS. I don't think about GIS much when thinking about remote sensing data, although some raster GISes do well with RS data sets and they do have mapping tools and so forth built in. I think the original posting is a red herring, there are a lot of relevant tools available already.
I am quite civilized, and I should be brought a beer immediately. -- Bruce Sterling
Another victim of the lethal GPL virus.
R.I.P.
"Free" remote sensing software has been floating around the academic community for decades. Its the image data that cost big bucks. Academic software isn't as polished as industry maintained stuff.
There is now a vast amount of remote sense data floating about the world.
...
The major types are satellite data ('photographic' and multiband), aerial ( photographic / geophysical ) and seismic ( marine / land ), with a plethora of 'spot' readings of various types.
The stages of data use are:
1) Aquisition - downlink from sat, fly plane & aquire, send ship out & measure, etc.
2) Georeference - correct perspective distortion / grid raw data , etc.
3) Delivery - Fat data over thin wire problem.
4) Process / Enhance / Interact with / Use.
In the world outside of mineral exploration / military use, the big news has been the Ikonos sat. & the explosion of much cheaper & higher resolution air photo mosaic's.
( One of the local councils has a 25cm air mosaic several Gigs in size & amoungst other uses, roofing tilers are grabbing the data to target people with tiles missing off their roofs )
One of the current hot topics is how to deliver terrabytes of raster/vector data over inet connections for people to use at home or in the office.
My current URL ( http://www.earthetc.com/ )( nope - I don't work for them, although friends do ) has a pretty impressive demo of zomming and panning over the net.
The OS community is looking at catching up with about thirty years of commercial / military software development stemming from USGS code, GRASS, GMT etc. Various components such as orthorectification have have dedicated commercial packages ( Halava , ermapper ) worth several thousand dollars.
An interesting tack for the OS movement to tack would be to leapfrog the aquisition & georeferencing stages, assume the existence of large georeferenced mosaiced data sets, and focus on the issues of delivery and end user tools, ( smart street directories, hand held's with 200Gb drives for use in co-ordinating emergency response teams etc.( requires integration with 'live' GPS data)).
It's a pretty interesting field
the resolutions would require new sattelites. The Continuous feeds would require either Geostationary sattelites or recieving towers spread out over the earth in a grid covering EVERYTHING... oceans and all.
Pat
(FIT Student)
Humans are slow, innaccurate, and brilliant; computers are fast, acurrate, and dumb; together they are unbeatable
To get that sort of resolution you would need enormous sats in orbit. Lots of sats. I don't think your voyeuristic tendancies are worth that much money.
I'm a loner Dottie, a Rebel.
GRASS IS GPLed for years now
Minor correction: version 5 of GRASS was released under the GPL last November (1999). GRASS v0 - 4.x is public domain (still under minor development, mostly bug fixing).