Difference between revisions of "RoboCup Vision"

From RoboJackets Wiki
Jump to navigation Jump to search
(added some basic vision information)
 
 
(7 intermediate revisions by 4 users not shown)
Line 1: Line 1:
The vision system handles all camera communication and outputs static vision pose information. It runs as a service as a [http://en.wikipedia.org/wiki/User_Datagram_Protocol| UDP] server that provides vision data for interested parties. When the vision service is started, it does nothing with the information it gathers until a client requests to have information sent to it. Once a client is "registed" with the service it will be sent new vision information when such information is available.
+
The vision system handles all camera communication and outputs static vision pose information. It runs as a service as a [http://en.wikipedia.org/wiki/User_Datagram_Protocol| UDP] server that provides vision data for interested parties. Once the vision service is started it begins to UDP broadcast the data for any interested parties to listen in on.
 +
 
 +
==Output Data==
 +
Basic requirements and layout of the output data that comes from the vision system.
 +
* Sequence # on each packet
 +
* Arbitrary # of objects
 +
* TODO: format
 +
 
 +
==Vision Processing==
 +
* End result also contains all intermediate data to ease debuging
 +
 
 +
==Vision Config==
 +
* masking zones
 +
* saving (separate files for different components ie. color tables)
 +
* Camera Setup (brightness, exposure, etc...)
 +
* Parameters
 +
** Intrinsic
 +
** Extrinsic
 +
* ''self calib''
 +
 
 +
[[Category: RoboCup]]
 +
[[Category: RC Software]]
 +
[[Category:2006-2007]]

Latest revision as of 20:25, 24 May 2020

The vision system handles all camera communication and outputs static vision pose information. It runs as a service as a UDP server that provides vision data for interested parties. Once the vision service is started it begins to UDP broadcast the data for any interested parties to listen in on.

Output Data

Basic requirements and layout of the output data that comes from the vision system.

  • Sequence # on each packet
  • Arbitrary # of objects
  • TODO: format

Vision Processing

  • End result also contains all intermediate data to ease debuging

Vision Config

  • masking zones
  • saving (separate files for different components ie. color tables)
  • Camera Setup (brightness, exposure, etc...)
  • Parameters
    • Intrinsic
    • Extrinsic
  • self calib