We tried to use the sample vision code that came from the latest c++ update and we were not able to get it to work. We placed debug lines to print out to the driver station lcd after every line and found the bad line of code.
What kind of checking are you doing to verify your image is valid when you call camera.getImage.
Last year, people had some issues with images that were valid pointers, but had 0 width, and would crash (due to an exception). I used this guard code after getting the image, but before converting to a binaryImage.
This line is giving us a height of 3670016. Does this seem like the correct height value for the image? We don’t think this is the right value for height. What do you think is causing this?