Product Q&A: Your Questions Answered

Dion · May 29, 2024, 5:43am

Hello everyone,

Recently, we have compiled a list of questions from customer feedback and have provided answers below. This will help other customers who might be facing similar issues find solutions quickly. If you have deeper insights into these questions, please feel free to share your thoughts in this thread. However, if your question is not listed here, please start a new thread.

This thread is active and regularly updated.

Dion · May 29, 2024, 6:10am

1. Would the vlc capture device not get a picture from video0 or video1?

Background:
The customer used an IMX219 MIPI camera on Raspberry Pi 5.

Dion · May 29, 2024, 6:14am

A:
We use libcamera instead of VLC to stream the MIPI camera on Raspberry Pi. When libcamera starts, it enumerates all available camera devices in the system. This involves scanning the /dev directory for potential camera device files like /dev/video0 .

Dion · May 29, 2024, 7:09am

2. What’s the chance there are some problems with the a single connector doing displays/camera’s both. (Zero2 & RPI 5)

Dion · May 29, 2024, 9:11am

A:
On the Pi 5, the camera interface and the display interface do indeed share the same connector, but configuration is required in the config file. However, the Pi 0 does not support this feature.

Dion · July 25, 2024, 1:57am

3. Are there any demo Python codes available for streaming cameras on the Jetson platform?

Edward · July 25, 2024, 2:02am

Please do not install the official opencv, nvidia has released a customized version of opencv

sudo pip uninstall opencv-python
sudo pip3 uninstall opencv-python

sudo apt install nvidia-opencv
sudo apt install nvidia-opencv-dev

You can refer to the following example:

github.com

ArduCAM/MIPI_Camera/blob/master/Jetson/IMX477/AF_LENS/JetsonCamera.py

# MIT License
# Copyright (c) 2019 JetsonHacks
# See license
# Using a CSI camera (such as the Raspberry Pi Version 2) connected to a
# NVIDIA Jetson Nano Developer Kit using OpenCV
# Drivers for the camera and OpenCV are included in the base image

import cv2
import time
try:
    from  Queue import  Queue
except ModuleNotFoundError:
    from  queue import  Queue

import  threading
import signal
import sys


# def signal_handler(sig, frame):

This file has been truncated. show original

Note: The code examples are not suitable for all cameras. The resolution, frame rate, etc. need to match the actual situation.

Dion · July 25, 2024, 8:32am

4. Cannot Install Camera Driver on Jetson Orin Nano/NX and Jetson AGX Orin

Edward · July 25, 2024, 8:34am

First,
Verify Carrier Board Compatibility: Please ensure that you are using an official NVIDIA carrier board. Our drivers are not compatible with third-party carrier boards. If you are using a third-party board, this may be the cause of the issue.
Update to the Latest L4T Version: Flash your system to the latest L4T version 36.3.0. After updating, try installing the driver again. The latest updates may have resolved compatibility issues.

Then,
If you still cannot install it, please update the kernel to the latest one.

sudo apt-get update
sudo apt-get upgrade

sudo reboot

After restarting the system, re-execute the camera driver installation command.

Dion · September 9, 2024, 3:17am

5. Cannot get compile.sh working on the Jetson device.

Dion · September 9, 2024, 3:20am

Please follow the steps below.

# Clone Tof Repository
cd ~
git clone -b v1.0.3 https://github.com/ArduCAM/Arducam_tof_camera.git
cd Arducam_tof_camera
# Install Dependencies
./jetson/Install_dependecies_jetson.sh
wget https://github.com/ArduCAM/arducam_ppa/raw/master/pool/main/a/arducam-tof-sdk-dev/arducam-tof-sdk-dev_0.1.3_$(dpkg --print-architecture).deb -O arducam-tof-sdk-dev_0.1.3.deb
sudo dpkg -i arducam-tof-sdk-dev_0.1.3.deb
# Compile
./compile.sh

Dion · September 23, 2024, 3:32am

6. Camera Not Detected in Web UI After System Update

Issue: Some users have reported that after performing a system update, the web UI fails to detect the camera. This issue typically occurs after running the following update commands:

sudo apt update
sudo apt upgrade

These system-wide updates, while generally beneficial, can sometimes lead to unexpected compatibility issues with existing hardware or software configurations. In this case, the camera detection functionality in the web UI appears to be affected.

Dion · September 23, 2024, 3:33am

Solution: Reflash the System Image

To resolve this issue, we recommend the following steps:

Reflash your device with the latest official image. You can download it from the following link: B0513 ArduCam 64MP Image (2024-05-22)
After reflashing, it’s crucial to temporarily avoid system updates. This will help maintain the current configuration that is known to work correctly with the camera.

Important Notes:

Reflashing will erase all data on the device. Make sure to back up any important files before proceeding.
The provided image is specifically configured to work with the camera and web UI.
We are actively working on a more permanent solution that will allow for system updates without affecting camera functionality.

Dion · September 25, 2024, 7:52am

Troubleshooting Guide: Image Capture Timeout with Arducam 64MP Quad Kit on Raspberry Pi 4 and 5

Problem Description

When using the Raspberry Pi (4 or 5) with the Arducam 64MP Quad Kit to capture images at maximum resolution, users may encounter a timeout error. This issue occurs when executing the following command:

libcamera-still -t 10000 -o test1.jpg --width 9152 --height 6944

The error message typically indicates a timeout during the image capture process.

Dion · September 25, 2024, 8:18am

Solution

The solution differs depending on whether you’re using a Raspberry Pi 4 or Pi 5.

For Raspberry Pi 5

Download the rpi_apps.yaml file from the provided link. rpi_apps.yaml - Google Drive
Place the rpi_apps.yaml file in the following directory:

/usr/share/libcamera/pipeline/rpi/pisp/

Set the following environment variable:

export LIBCAMERA_RPI_CONFIG_FILE=/usr/share/libcamera/pipeline/rpi/pisp/rpi_apps.yaml

To make this change permanent, add the above line to your ~/.bashrc file:

echo 'export LIBCAMERA_RPI_CONFIG_FILE=/usr/share/libcamera/pipeline/rpi/pisp/rpi_apps.yaml' >> ~/.bashrc

Apply the changes using the source command:

source ~/.bashrc

For Raspberry Pi 4

Edit the following file:

/usr/share/libcamera/pipeline/rpi/vc4/rpi_apps.yaml

Add the following line to the file (ensure to include the comma):

"camera_timeout_value_ms": 50000,

Save the file.
Set the following environment variable:

export LIBCAMERA_RPI_CONFIG_FILE=/usr/share/libcamera/pipeline/rpi/vc4/rpi_apps.yaml

To make this change permanent, add the above line to your ~/.bashrc file:

echo 'export LIBCAMERA_RPI_CONFIG_FILE=/usr/share/libcamera/pipeline/rpi/vc4/rpi_apps.yaml' >> ~/.bashrc

Apply the changes using the source command:

source ~/.bashrc

Applying Changes

After making these changes, you need to apply them:

If you’ve edited ~/.bashrc manually, run:

source ~/.bashrc

Alternatively, you can restart your Raspberry Pi to ensure all changes take effect:

sudo reboot

Verification

After applying the changes, try running the image capture command again:

libcamera-still -t 10000 -o test1.jpg --width 9152 --height 6944

The timeout issue should now be resolved, allowing you to capture images at maximum resolution without errors.