Posts Tagged google
Google street view dissection
I felt like working out how google street view does its stuff today – so here goes!
Here’s the image of the street view interface we’re dealing with
Time to open up wireshark and see how it gets these images!
Looks like there’s lots of tile requests to the server at cbk2.google.com
GET /cbk?output=tile&zoom=3&x=4&y=1&cb_client=maps_sv&fover=2&onerr=3&v=4&panoid=0JTH3YvHt93HUiWezPIRhgHTTP/1.1\r\n
Pats of interest are the zoom, x, y, and panoid tags. I presume x and y are the tile locations in the entire panorama and zoom is the zoom level (0 to 3 in this case). panoid must be a specific identifier for the street view image you’re looking at.
putting this into a browser retrieves this tile
click image to go to the google version
Nice! now for a play – Lets change the x and y values to (0,0)
mmm.. don’t know where that is. Lets look down a little at (0,1)
ahh! There’s something interesting. Still, 0 on the x axis doesn’t seem to be “forward”. Lets look at the tile to the left which is at (6,1)
That explains it… Looks like there’s some overlap at the join. And – taking a closer look at this tile shows that the join is actually on the “back” of the image (as it was captured) because there’s a one way arrow on the ground. Perhaps tile 0 is actually aligned to west? Not sure on this one.
Ok, time to play with zoom. Here’s zoom=4 at (0,0)
A value of 4 doesn’t seem to work. It just gives a black tile. I think that’s because this area (in Perth WA) was only covered at a fairly low resolution this time around (it looks like they used ladybug2 cameras for this pass).
Well, here’s zoom = 2 at (0,0) then
That’s more like it. Interestingly, the join point on the x axis is still the same (i.e. the left of tile 0).
Now for zoom =1
and zoom=0
So there we go.
Just for fun I stitched all the zoom=1 tiles to generate this
Definitely looks like a Equirectangular_Projection
Out of interest I thought I’d also estimate the total size for each full sized image to be about 750kB. This assumes a tile size of 32kB, 21 full tiles and 7 quarter tiles at the bottom. Say that’s 1MB per image including the lower scales, multiplied by the number of streetview images over the world…. equals a lot!
Google WiFi data collection
Google have caused quite a FUSS after their accidental collection of unencrypted WiFi data.
“Communications Minister Stephen Conroy has lashed out at Google, accusing the internet giant of the single biggest breach of privacy in history.”
Here’s the official Google response.
I posted an article in February last year on how/why the street view cars collect WiFi data and I think it’s quite a clever idea. Basically, Google holds a database of WiFi MAC addresses collected by the street view car and users of their Latitude application. This database is then used to provide location information to users without a GPS device (i.e. on your laptop).
Google aren’t denying they collect WiFi access point information. Their mistake was storing un-encrypted WiFi data. Every WiFi access point broadcasts information about itself including a unique identifier (the MAC address) which is what google want. WiFi also broadcasts all data which is transferred – such as internet banking passwords and emails – meaning anyone within a reasonable distance can collect this private information. To prevent this most WiFi links use encryption to scramble the data making it unintelligible to anyone who doesn’t know a secret password. It wouldn’t matter if Google collected encrypted WiFi data because they can’t decrypt it.
So really, if you use an unencrypted WiFi link you really only have yourself to blame. An analogy of unencrypted WiFi communications would be sitting in your house with the windows open yelling your internet banking details over a megaphone for all your neighbors to hear. An encrypted WiFi link is slightly better because you’d be yelling the details in your own made-up language. I think Google are the least of your concerns.
It will be interesting to see what comes of the various government investigations into Google’s privacy breech. Maybe here in Australia Steven Conroy could add google.com to the proposed internet filter.
Programming for dummies
I just stumbled across Scratch.
Ok, it’s meant to be for kids, but I like the idea. It’s basically a visual programming environment for creating animation and music.
I found this via a post on the Google research blog about “Android app inventor“. They’re planning to make a similar graphical block-based programming environment for writing android apps.
I like the idea of visual programming. The traditional code-compile-run-test-stop-change-compile-run sequence is a little boring and time consuming. I want to change something “on the fly” and see what it does.
This programming style often seems to be associated with education. Greenfoot and Alice are two examples. If I was a real engineer, I may find myself playing with LabVIEW for programming microcontrollers and whatever else real engineers do.
But anyway. A couple of years ago I started working on a visual programming environment for creating video/audio/frame based processing applications – FraMMWorks. The idea is that you have a collection of building blocks (video sources, face detection, edge detection, etc) which you join together to generate an overall video processing “chain”. You can then save the resulting code as an executable and run it anywhere.
If there’s any interest I might just spend some more time on it.
Google Real-estate
Browsing through this morning’s articles I came across a blog post for the new Google Maps Real Estate layer.
It’s a really nice visulisation of what is currently for sale/rent. And the best part – it works for Perth!
Here’s an example for some typical “first-home-buyers” in the Perth area. It’s interesting to see clusters north, south and east and some areas with no houses at all.
Energy efficiency Microsoft style.
Microsoft’s latest technology – Hohm – is ready to save the world from global warming.
It sounds, ugh, confusing.
The general concept behind it is good. In fact, it seems similar to Google’s offering – PowerMeter – which borrows the catchline:
If you cannot measure it, you cannot improve it. — Lord Kelvin
I can’t see how these “applications” will work yet though. It all seams a bit abstract and idealistic. The main issue is that someone needs to make a buck. Who wants to pay? The energy providers? Unlikely – they’ll loose money through reduced energy usage anyway. End users, well yea, but why would I want to pay for something I already know?
I suppose in essence it’s really about re-educating people. Which means it’s not going to make anyone any money. If companies such as Microsoft and Google have a spare few mil to throw at this – good on ’em. But it seems a little fishy to me.
And for something amusing:
“Later on, Microsoft intends for the service to grab data from programmable thermostats and so-called “smart plugs” to provide better real-time information. It all sounds well and good until the day we’re inevitably forced to call Microsoft’s activation center after adding a deck to the house”
http://www.theregister.co.uk/2009/06/24/microsoft_launches_hohm/
Google Latitude + GoogleCar
Everyone has heard about Latitude by now. “See your friends’ locations and status messages and share yours with them.”
Ok, there’s a mountain of privacy issues surrounding it… Simple solution — turn it off. I have no intention to broadcast my location 24 hours a day. That’s just asking for trouble. But it could be very handy when you actually want your location to be known.
Anyway – what I’m interested in is how it finds your location?
Most smart phones have some sort of Location API – Here’s an example using the Symbian Location Acquisition API. Java and iPhone have similar API calls.
The other side of the API presumably uses whatever it can to get a location. GPS (possibly with the help of an A-GPS server) or some form of mobile network based “triangulation” seem to be popular.
But what if your phone doesn’t have a GPS receiver? Or you’re using the Latitude iGoogle app on your laptop? Or a desktop?
This is where things get interesting. The Latitude iGoogle app gets your location through the Google Gears GeolocationAPI which according to google, uses “your computer’s WiFi location”. What the hell is your “WiFi location”?
Well, if you can build up a big enough database of WiFi access point MAC addresses and associate each with a location, you can perform a “reverse” query to associate an AP to a location! viola! Each AP should have a 100m (or so) transmission radius, so if there are multiple APs you can “overlay” these transmission circles to get an even more accurate position.
But this theory only works if you have a lot of entries in your database and it’s regularly updated (as people add/remove/move their personal APs). How do you do this? My ideas:
- The Google street view car scans for WiFi signals as it drives around
- Every Laitude program updates the Google database while it is being used
The Street view car visits every street which gives Google a good starting point for the database. As soon as possible after this they get users to “give Latitude a try” which corrects the database with any changes since the Google car went past, making it more accurate as more users come on. To back up this Google car collection claim, I’m going to use a quote from the official google blog “Today, Latitude is available in 27 countries, and we hope to add more soon.”. How many countries do you think Google has Street View imagery for or is currently capturing? About 27 perhaps?
It’s all very clever if you ask me.
Just don’t tell anyone what Virtual Observer is currently doing with rubbish trucks…… 😉
Recent Comments