Difference between revisions of "Team Commando"
From GcatWiki
(→To Do) |
(→To Do) |
||
Line 6: | Line 6: | ||
**https://docs.google.com/document/d/1YrS9Ua821oJdYNquYGewYql7AsK0MqUhzOm-0q8JxM8/edit | **https://docs.google.com/document/d/1YrS9Ua821oJdYNquYGewYql7AsK0MqUhzOm-0q8JxM8/edit | ||
*poster | *poster | ||
+ | **[[File:TeamCommando.ppt]] | ||
==Navigation== | ==Navigation== |
Revision as of 00:57, 2 May 2012
To Do
- command module navigation (Daniel and Matt)
- image processing (Steph, Evan, Jonah)
- decide on a camera
- abstract
- poster
Image Processing
- camera
- simple webcam!
- software
- matlab toolbox
- http://homepages.inf.ed.ac.uk/rbf/BOOKS/PHILLIPS/cips2ed.pdf
- Roborealm
- method
- attach laptop to iRobot
- receive signal from robot when it has arrived in room
- use laptop to take image
- process image to determine whether or not a window is open
- Current dataset
- 26 closed windows
- 26 open windows
- https://www.dropbox.com/sh/hx5rwp9ilakwg5c/8Yf4twUjV1
- RoboRealm
- Color to detect: magenta?
- attach color to bottom of window so can only be detected when open
- to-do:
-figure out physical orientation of computer/camera
-build training set: every possible window configuration
-figure out how to get the images... opencv documentation: http://opencv.willowgarage.com/documentation/cpp/features2d_common_interfaces_of_feature_detectors.html, http://opencv.willowgarage.com/wiki/CameraCapture
-decide on preprocessing method: threshhold, line detection, contrast upping, pulling out the windows, or doing none of that and trusting the neural net.
-figure out how to get the command module to tell comp when to take a picture
-get the neural net trained and tested
- Active Thoughts:
-RoboRealm seems sweeeeell.
- interface with MATLAB: http://www.mathworks.com/matlabcentral/fileexchange/22882
-current plan-ish:
- roborealm does processing and recieve signals from command module via serial port, then sends processed images ot matlab for NN classification. BOOM.