Best Possible Configuration for Capturing Image

Developer Tools Other Tools
, , , ,
1

Hi,
I’m trying to capture an image as quickly as possible using a pi camera attached to the Jetson Nano and then process it.
Currently I use this parameter to capture the image in Python

dispW = 320
dispH = 240
flip = 2
camSet='nvarguscamerasrc !  video/x-raw(memory:NVMM), width=3264, height=2464, format=NV12, framerate=21/1 ! nvvidconv flip-method='+str(flip)+' ! video/x-raw, width='+str(dispW)+', height='+str(dispH)+', format=BGRx ! videoconvert ! video/x-raw, format=BGR ! appsink'

I calculated the time before starting the gstreamer and after grabbing an image. It takes roughly 1.2 sec to write the image to a destination folder. The time is quite high for me. An option is to maybe reduce the resolution of the camera. Time consumed by cv2.VideoCapture() is nearly 1 sec but are there any possibilities of maybe writing an image faster like:

  1. Run the OpenCV program on GPU instead of CPU
  2. Change GStreamer config
  3. Use other programming languages like C++ or GoLang
  4. Or use a USB3 camera for faster image processing

Any suggestions?