stereo_image_proc failing for Prosilica cameras when binning [closed]

edit

Hi,

We have a pair of Prosilica 2450C GigE cameras that we were trying to calibrate and use with stereo_image_proc---specifically for computing the disparity map and corresponding 3d point cloud.

Trying to do this using full resolution (they are 5MP cameras) created some problems because of limited bandwidth, and a small window (compared to the image size) for the disparity computation (128 pixels). We got an output, it just was very sparse (and slow), not entirely surprisingly.

Instead, we tried to increase the binning of the camera (specifically using a binning value of 4 for both X and Y and resulting in a 612x512 image). We will lose color for these cameras when we do this, but that's not a huge concern right now for us. However, when trying to connect to the disparity topic, we received an error message related to image size mismatch:

$ ROS_NAMESPACE=stereo rosrun stereo_image_proc stereo_image_proc
[ WARN] [1349133972.827883520]: Color topic '/stereo/left/image_color' requested, but raw image data from topic '/stereo/left/image_raw' is grayscale

OpenCV Error: Assertion failed (0 <= roi.x && 0 <= roi.width && roi.x + roi.width <= m.cols && 0 <= roi.y && 0 <= roi.height && roi.y + roi.height <= m.rows) in Mat, file /tmp/buildd/ros-fuerte-opencv2-2.4.2-0precise-20120908-1632/modules/core/src/matrix.cpp, line 322 terminate called after throwing an instance of 'cv::Exception'
  what():  /tmp/buildd/ros-fuerte-opencv2-2.4.2-0precise-20120908-1632/modules/core/src/matrix.cpp:322: error: (-215) 0 <= roi.x && 0 <= roi.width && roi.x + roi.width <= m.cols && 0 <= roi.y && 0 <= roi.height && roi.y + roi.height <= m.rows in function Mat

Aborted (core dumped)

Any body else run into this?
Thanks!