I want to develop a code that does the following task:
open a camera device e.g. /dev/video0
configure device: exposure time, gain, …
stream the result to a second process (e.g. shmsink/udpsink) for processing
What i the best practice?
Or is there an example somewhere?
Thanks for help
DaneLLL
November 29, 2019, 2:23am
#2
Hi,
You may try rtsp in gstreamer and test-launch is easy to use. Please refer to
https://devtalk.nvidia.com/default/topic/1018689/jetson-tx2/vlc-playing-gstreamer-flow/post/5187270/#5187270
Also please share information about your camera. For Bayer sensors, we use nvarguscamerasrc. For USB cameras or YUV sensors, we use v4l2src. Using which plugin depends on the camera types.
Hi,
I use an Sony IMX390 sensor connected via FPDLinkIII SerDes. The carrier is an own developed board.
DaneLLL
December 2, 2019, 2:06am
#4
Hi,
IMX390 looks to be Bayer sensor. Please check if you can run
$ gst-launch-1.0 nvarguscamerasrc ! nvoverlaysink
If sensor driver and device tree are programmed accordingly, you should be able to launch preview in executing the command.