Hi everybody,
I'm having some trouble to make the last version (I had no problem with previous versions) of v4l4j working with the Intel QX3.
The problem is that if I use a JPEGFrameGrabber as I always did (and which I'm sure that is supported by the device because I tested it with videoDevice.supportJPEGConversion()), I get images as the one attached. Image that is clearly not from the device stream.
Here the code used:
int std = V4L4JConstants.STANDARD_WEBCAM;
int channel = 0;
int quality = V4L4JConstants.MAX_JPEG_QUALITY-10; //If I use the max it crashes after 4/5 seconds
[ . . . ]
frameGrabber = videoDevice.getJPEGFrameGrabber(640, 480, channel, std, quality);
In the other hand if i use the native image format that is RGB24 (tested with frameGrabber.getImageFormat()) a null pointer exception is threw.
Here the output:
Intel Play QX3 Microscope on /dev/video1
[ v4l2-input.c:255 ] CAP: VIODIOC_S_FMT succeeded but returned a different palette from the one requested: requested palette 0x33424752 - returned palette: 0x41495043
[ v4l2-input.c:256 ] This is most likely a bug in v4l4j. Please
[ v4l2-input.c:257 ] let the author know about this issue. See README file.
[ v4l2-input.c:405 ] Unable to set the palette: RGB24
[ v4l2-input.c:406 ] Please let the author know about this error.
[ v4l2-input.c:407 ] See the ISSUES section in the libvideo README file.
[ v4l2-input.c:606 ] Listing the reported capabilities:
Error in MainFrame.initFrameGrabber()
Message:
Image format RGB24 not supported
Cause:
null
Exception in thread "main" java.lang.NullPointerException
Does anybody have an idea?
Have a nice day you all,
Nicola.