I had similar stuff happen to me with a Philips SPC1300 on our
humanoid robots. Approx every 500th frame had corruption that looked a
lot like some parts of the image data array just went missing. IIRC we
exchanged the USB cable and it worked again. It definitely was a
hardware issue in our case.

Sample image:
http://img96.imageshack.us/img96/2705/cammosaicluise145.png

2011/2/12 Antons Rebguns <anton@email.arizona.edu>:x
> Chris,ch
>
>> The problem does not happen in smallv and does not appear to happen in
>> coriander.  It is present in disparity view when using image_view's
>> stereo_view node.  videre_stereo_cam's disparity_view displays no disparity
>> image.
>
> I fixed the disparity display, apparently the time stamp on disparity
> image was missing that is why the synchronizer couldn't get two
> matching images and did not display anything. I am still not sure what
> the problem might be. I just plugged in STOC that we have here and it
> works just fine. The firmware versions are newer though, but I doubt
> that would cause this problem:
>
> Camera Firmware: 07.00
> Imager Firmware: 0007
> STOC Version: 04.02
>
> I'll keep looking, maybe something jumps at me as I look at the code.
>
> Anton
>
>
>> Here's some info about the camera in case it matters:
>>
>> Camera Firmware: 06.02
>> Imager Firmware: 0006
>> STOC Version: 04.02
>>
>> Thanks,
>> Chris
>> On Fri, Feb 11, 2011 at 10:14 AM, Antons Rebguns <anton@email.arizona.edu>
>> wrote:
>>>
>>> Chris,
>>>
>>> That's a very strange problem you have there. Does the same mangling
>>> happen in point cloud received from the device? Does it happen in when
>>> you run the driver in non-STOC mode? Do you see the same problem if
>>> you run the camera in coriander, for example?
>>>
>>> Thanks,
>>> Anton
>>>
>>> On Thu, Feb 10, 2011 at 8:48 PM, Christopher Bessent <cmbq76@mst.edu>
>>> wrote:
>>> > Hi all,
>>> > My group is attempting to use one of Videre's STOC cameras with the
>>> > videre_stereo_cam node from ua-ros-pkg.  It works fabulously except for
>>> > the
>>> > bottom third which appears to repeat the upper right corner of the
>>> > respective image.  Every once in a while that section will also devolve
>>> > into
>>> > complete static.
>>> > Here's some example images:
>>> > http://i.imgur.com/g485H.png
>>> > http://i.imgur.com/J1UO3.jpg
>>> > The only idea I have left is that we haven't made a
>>> > calibration/rectification for this camera yet, but I can't imagine that
>>> > causing this kind of problem.
>>> >
>>> > --
>>> > Chris Bessent
>>> >
>>> >
>>> > _______________________________________________
>>> > ros-users mailing list
>>> > ros-users@code.ros.org
>>> > https://code.ros.org/mailman/listinfo/ros-users
>>> >
>>> >
>>
>>
>>
>> --
>> Chris Bessent
>> cmbq76@mst.edu
>> Missouri S&T Undergrad Computer Engineering
>> Eta Kappa Nu - Bridge Correspondent
>> Robotics Competition Team - Computing Lead
>>
>>
>> _______________________________________________
>> ros-users mailing list
>> ros-users@code.ros.org
>> https://code.ros.org/mailman/listinfo/ros-users
>>
>>
> _______________________________________________
> ros-users mailing list
> ros-users@code.ros.org
> https://code.ros.org/mailman/listinfo/ros-users
>