Image processing data collection help /advice

Hello All,

I have the Jetson TX2 attached to a camera with it trained to detect certain images. Does anyone know how if there’s a way to stream the data it picks up/ location (coordinates on the screen) to a database?
and what protocol it uses.

Thank you!

hello ilovechipmunks,

may I have more details. did you meant crop the specific region from the sensor stream?

if yes,
you may running with gst pipeline to add video converter with crop region (left, right, top, bottom)
for instance,
$ gst-launch-1.0 nvarguscamerasrc sensor-id=0 ! 'video/x-raw(memory:NVMM),width=2592, height=1944, framerate=30/1, format=NV12' ! nvvidconv left=300 right=2200 top=100 bottom=1800 ! xvimagesink