The following link is a news article about a project called “Large Synoptic Survey Telescope”. The $400-million project tries to build a telescope to survey the entire sky every three nights for ten years. Honestly this is not the first “science” project that has its way to waste money in history. Nowadays people just obsessed with collecting data, any kind of data from any sources; indeed the biggest problem is what to do with all the data? Unfortunately Hollywood always tells the public that it is quick and easy to find, say, a person from various surveillance cameras but this is a huge job indeed. If it is so easy to do, Mr. bin Laden won’t be still at large at this time. All these imageries require analysts to examine inch-by-inch before one photo is found to be useful. The problem with the American military is not they don’t have enough airborne surveillance platform but they don’t have enough eyes to analyze the imagery data. (Infrared, SAR or hyperspectral imageries even require trained eyes.) In the case of astronomy, I believe most discoveries by using Hubble Telescope came from imageries that were taken a long while ago. It takes time and expertise to examine these photos. What good does it do to stream the imageries of the entire night sky to the internet for the general public? Yes, perhaps we now have the technology to do this kind of survey but we must think beyond achieving the technology milestone.
Surveying the night sky is quite different from airborne surveillance of the earth surface, I think. The night sky is pretty much a constant set of images and now they have the image digitized, which you can let the computer to do the comparison. In the old days, astronomers have to work on films with their own eyes. The significance of that project is that if astronomers look at the whole sky frequent enough, they can detect small and short-cycle changes. That should help detecting tiny asteroids and blackholes. Computer recognition is also entering security surveillance. (even my Canon camera has "face detection" function...). I saw a news report scientist now use normal surveillance cam combining with IR cam, and let the computer to do the face recognition. It also analyse the walking gesture of the subject. If you want to trace an individual, given you hv enough sets of those surveillance cams and computing power (that means $$), you can certainly do it.
回覆刪除[版主回覆01/07/2008 07:14:00]Yeah, you are right, this kind of surveillance is different from security surveillance. I thought they were trying to look at deep space instead of searching for small asteroids. Now when I think about it, the method is suitable for looking for asteroids. (Not black holes; black holes have different signature, land-base observation is not effective at all.)
There is a phenomenon called "gravitational lens", which is caused by a massive object (e.g. blackholes, neutron stars...) passing between the observer (in this case, we on earth) and distant star light. The gravitational pull would bend the light path and cause a shift in the star image. I guess the programme may shed some light on those objects, which themselves giving little or no light at all.... see: http://en.wikipedia.org/wiki/Gravitational_lens
回覆刪除[版主回覆01/08/2008 10:48:00]For observation of this nature, they should pinpoint to an area of interest instead of taking pictures of the whole sky and hope to find something interesting. In security surveillance, a positive ID is the goal but in astronomical research, the goal is a negative ID. That is to say they are looking for something they don't have in the database. This is really different.
Gravitational lensing, that's something interesting!