[OSM-talk] Laser scanning (Video

Andy Robinson Andy_J_Robinson at blueyonder.co.uk
Fri Mar 31 14:50:16 BST 2006


This was my thought originally in gathering photos at junctions and major
bends. In that coupled with waypoints it could be used to quickly create a
rough first draft map. In the end I found a first pass logging and second
pass additional data gathering on foot was a better approach for me.
Certainly more interesting that just documenting junctions and bends.


Andy Robinson
Andy_J_Robinson at blueyonder.co.uk 

>-----Original Message-----
>From: talk-bounces at openstreetmap.org [mailto:talk-
>bounces at openstreetmap.org] On Behalf Of Tom Carden
>Sent: 31 March 2006 14:27
>To: Talk Openstreetmap
>Subject: Re: [OSM-talk] Laser scanning (Video
>On 3/31/06, Daniel Haran <chebuctonian at gmail.com> wrote:
>> Since we're talking about automatically acquiring and recognizing
>> data, one thing I've wondered about is whether streets signs could be
>> automatically "read"/parsed, along with house numbers and traffic
>> signs. Does anyone here know if the technology is getting advanced
>> enough?
>This is something I'd like to try, but I reckon that 'mechanical turk'
>approaches would be best - e.g. for the IoW I'd like to try using
>Flickr with triple-tags which could be used to annotate OpenStreetMap.
> For instance if a photo is geotagged with geo:lat=y and geo:long=x
>and streetnametagged with streetname=blah, then we could use this to
>set the name of the nearest street segment in OpenStreetMap, within
>some threshold (e.g. 10m).
>There are plenty of edge cases where street signs aren't 100% accurate
>for the actual name of the street though:
>talk mailing list
>talk at openstreetmap.org

More information about the talk mailing list