Robot Navigation by Stereoscopic Camera

Robot Navigation by Stereoscopic Camera uses OpenCV library to compute disparity map and depth map from a pair of images taken by pair of two cameras. System can also detect nearest objects in scene, compute direction and angle to them...
In the end, robotic platform is programmed to maintain constant distance from the nearest object.
System runs under Linux Fedora.

Category:

Howto & Style

Tags:

• robot
• navigation
• stereoscopic
• camera
• robotic
• platform
• stereo
• opencv
• image
• processing

[VanDucTech]

Hi yaaaratheone,

Can you let me know which function to extract the nearest Object.?

Thanks so much!

[yaaaratheone]

@VanDucTech there is not a special function for the nearest object extraction. I had to write it by my own. There were many steps from computing disparity map, using histograms to find similar distances, removing noise, etc...

[half4team]

Hiw did you synchronize the camera's?

[vinhdhbackhoa]

it's really great. I am doing a project like this, but they are only the beginning, I've only read the depth map from two cameras ...

to be like you, I need to make more material any more, if you can share your documents or code? I know you write in linux OS, can I do the same on Windows or not? I am very grateful for your help

[GIPetrou]

Nice work!!! How do you distinguish between different objects when you have the depth map? Is there a source code available for this?

[chosclonio]

Hi, I would like to know the combination of parameters of the function of cvFindStereocorrespondenceBM.

[csdaqing]

Are you using the BM algorithm of openCV to compute disparity？