Ask Your Question
0

Technique to detect a hospital bed by camera

asked 2013-01-24 00:55:19 -0600

this post is marked as community wiki

This post is a wiki. Anyone with karma >50 is welcome to improve it.

I'm want to detect a bed in room. One may be edge detection or may be region area defining. Please, help me to start with this.

edit retag flag offensive close merge delete

Comments

what are the different method s to detect a bed in a room with video camera through opencv 2.4.3 in c.one may be edge detection or may be region area defining..please tell any one method with example..

neerat.sharma gravatar imageneerat.sharma ( 2013-01-24 02:18:26 -0600 )edit

one i knew canny and hough transformation. other then it.

neerat.sharma gravatar imageneerat.sharma ( 2013-01-24 05:46:14 -0600 )edit
1

Some example images will help to understand your problem. It is hard to help you without that because no general solution for object detection exists.

Michael Burdinov gravatar imageMichael Burdinov ( 2013-01-30 00:24:52 -0600 )edit
neerat.sharma gravatar imageneerat.sharma ( 2013-01-30 01:58:42 -0600 )edit

pls help...can we apply here SIFT technique

neerat.sharma gravatar imageneerat.sharma ( 2013-01-30 01:59:39 -0600 )edit

I don't think SIFT will help you much, because each bed may appear completely different in each image: different angle, different illumination, sheets, pillows, or other objects on the bed. And so on and so forth. There will be very few points (if any) with similar SIFT features. It seems like an extremely complicate problem, unless you have some prior knowledge that may help in detection.

Michael Burdinov gravatar imageMichael Burdinov ( 2013-01-30 05:20:54 -0600 )edit

@Michael Burdinov thanks for atleast replying.........if i go with pattern recognition..how it will help some idea.

neerat.sharma gravatar imageneerat.sharma ( 2013-01-30 05:50:43 -0600 )edit

??????????

neerat.sharma gravatar imageneerat.sharma ( 2013-01-30 23:02:20 -0600 )edit

1 answer

Sort by ยป oldest newest most voted
1

answered 2013-02-13 10:02:46 -0600

this post is marked as community wiki

This post is a wiki. Anyone with karma >50 is welcome to improve it.

I think one of the best approaches is to use DPM.
It is the state of the art algorithm for object detection since it's publish date.

If speed is one of your concerns use HOG feature and a linear SVM.

edit flag offensive delete link more

Comments

Thanks a lot Man.

neerat.sharma gravatar imageneerat.sharma ( 2013-02-13 22:37:58 -0600 )edit

Please mark the answer as correct (click the tick mark) if you found it useful

Mahdi gravatar imageMahdi ( 2013-02-14 04:28:50 -0600 )edit

Question Tools

Stats

Asked: 2013-01-24 00:55:19 -0600

Seen: 802 times

Last updated: Feb 22 '13