Ask Your Question

Revision history [back]

click to hide/show revision 1
initial version

Tricky image segmentation in Python

I'm trying to segment a live video feed in real time. The frames look much like this: http://img515.imageshack.us/img515/9531/depthimg2.jpg. In particular, I need to extract pieces like the smaller box in the center-right of the frame.

The problem is that the background isn't completely black. I've been trying to use FindContours, but it has a hard time because it treats the image as binary. Is there some sort of blob detection that handles multiple layers?

Tricky image segmentation in Python

I'm trying to segment a live video feed in real time. The frames look much like this: http://img515.imageshack.us/img515/9531/depthimg2.jpg. http://img515.imageshack.us/img515/9531/depthimg2.jpg In particular, I need to extract pieces like the smaller box in the center-right of the frame.

The problem is that the background isn't completely black. I've been trying to use FindContours, but it has a hard time because it treats the image as binary. Is there some sort of blob detection that handles multiple layers?