loading
 
seeking suggestion
Peter from India  [1 posts]
4 years
I am completely new to roborealm so pardon me if I sound stupid. I am doing a project (artificial vision guidance for blind through acoustics). I am using a camera module to identify the obstacles . could you please suggest me how  it  is possible to recognise the obstacles for ex.human , tress , vehicles etc using softwares
Steven Gentner from United States  [1370 posts] 4 years
Peter,

The best generic way to detect obstacles is using some form of depth sensor. Inside you can use the Kinect and its variations but outside you will be limited to stereo or lidar based systems. The Kinect is very easy to use but since it uses IR light sunlight will cause it to fail. Stereo is well researched and good for navigation but may not have the precision and resolution that you may need for an individual walking.

Given that you probably need a small device that people can carry Stereo vision may be the way to go. Have a look at the Point Gray Bumblebee cameras and see if that is something you can afford for your project.

STeven.

This forum thread has been closed due to inactivity (more than 4 months) or number of replies (more than 50 messages). Please start a New Post and enter a new forum thread with the appropriate title.

 New Post   Forum Index