Problem Pipeline CSI camera Jetson Nano OpenCV

Autonomous Machines Jetson & Embedded Systems Jetson TX2
, ,
1

Hello,

I have some troubles to make my CSI camera work with OpenCV on Python.

I have already posted a similar topic but it was previously a thread problem that Nanocamera module fixed :

I upgraded Jetpack from 4.5 to 4.6 because it has native drivers to use IMX477 CSI cameras which I need for a project.
Now with the same python program as my previous post, I am not able to connect to CSI camera anymore.
I am able to make a pipeline on the console with the gst-launch-1.0 command and the same with nvgstcapture-1.0 command but I can t connect to CSI camera with Open CV on python (I have tried with a IMX477 camera and the Pi camera V2).
I tried with an USB camera and it works fine because I am using the command “cap = cv2.VideoCapture(1)”, not the pipeline.

I think something changed with the new version of OpenCV or Jetpack 4.6 and the pipeline command has to be modified a bit to work but I don t know how


Does anyone knows the solution to my problem ?

Thank in advance !

3

Hi,
Generally we run gstreamer pipeline in cv2.VideoCaprure() to use nvarguscamerasrc plugin. Please try the samples:
Opencv nvarguscamerasrc: hard LOCKUP on cpu 0 - #3 by DaneLLL

4

Hello !

Thank you very much for your reply

Sadly it doesn’t work

I have my IMX477 on the port /dev/video0

In the console :

$ gst-launch-1.0 nvarguscamerasrc ! 'video/x-raw(memory:NVMM), width=(int)1920, height=(int)1080,format=(string)NV12, framerate=(fraction)30/1' ! nvoverlaysink

works fine.

But in Python :

import sys
import cv2
def read_cam():
    cap = cv2.VideoCapture("nvarguscamerasrc ! video/x-raw(memory:NVMM), width=(int)1920, height=(int)1080,format=(string)NV12, framerate=(fraction)30/1 ! nvvidconv ! video/x-raw, format=(string)BGRx ! videoconvert !  appsink")
    if cap.isOpened():
        cv2.namedWindow("demo", cv2.WINDOW_AUTOSIZE)
        while True:
            ret_val, img = cap.read();
            cv2.imshow('demo',img)
            cv2.waitKey(10)
    else:
        print( "camera open failed")

    cv2.destroyAllWindows()


if __name__ == '__main__':
    read_cam()

doesn’t work.
It only print : camera open failed

I tried with all pipelines you have sent me
I used this one on jetpack 4.5 and it worked fine :

"nvarguscamerasrc ! video/x-raw(memory:NVMM), width=(int)1920, height=(int)1080, format=(string)NV12, framerate=(fraction)30/1 ! nvvidconv flip-method=2 ! video/x-raw, width=(int)1920, height=(int)1080, format=(string)BGRx ! videoconvert ! video/x-raw, format=(string)BGR ! appsink"

but now in 4.6, it doesn’t work anymore


5

If I launch the program from the console, this is the error message I get :

~$ python3 camtest.py
[ERROR:0@0.449] global /io/opencv/modules/videoio/src/cap.cpp (164) open VIDEOIO(CV_IMAGES): raised OpenCV exception:

OpenCV(4.5.5) /io/opencv/modules/videoio/src/cap_images.cpp:253: error: (-5:Bad argument) CAP_IMAGES: can’t find starting number (in the name of file): nvarguscamerasrc ! video/x-raw(memory:NVMM), width=(int)1920, height=(int)1080,format=(string)NV12, framerate=(fraction)30/1 ! nvvidconv ! video/x-raw, format=(string)RGBA ! appsink in function ‘icvExtractPattern’

camera open failed

6

Hi,
From the log it looks like the OpenCV is not built with gstreamer being enabled. You may rebuild OpenCV 4.5.5 and try again. Or re-flash the system to use default OpenCV 4.1.1

1 Like
7

Indeed it looks like Gstreamer is not activated in my OpenCV version

It looks pretty complicated to rebuid OpenCV so I will try to reinstall Jetpack 4.6 on the SD card and reinstall everything

Thank you again for your help

I will tell you if it worked or not

8

I have finally finished to re flash the SD card (update and upgrade everything and install pytorch) and everything works perfectly now !

I can use my previous python program with the IMX477 camera

Thank you so much for your help ! :)

1 Like
11

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.