Introduction
This Arducam quadrascopic camera bundle incorporates four 1MP OV9281 global shutter cameras, and the cutting-edge MIPI stereo camera breakthrough from Arducam – The Arducam Camarray HAT. As the upgraded version of Arducam Synched Stereo Camera HAT, the Camarray further pushes the multi-camera limits and doubles the interfacing capabilities to allow 4 synchronized MIPI camera modules to be interfaced to a single MIPI CSI-2 slot on prevailing single-board computers like Raspberry Pi, Jetson Nano, and Jetson Xavier NX.
Also, as the Camarray HAT releases, Arducam starts to support a variable baseline for stereo camera applications with some modified Arducam camera modules, such as those OV9281 ones included in this bundle. Since no shared camera board or clock-sharing soldering is required for the image sensors, you can kick off your prototyping for stereo camera applications and then decide the optimal baseline. Arducam also accepts subsequent customizations for your stereo camera applications.
What is Camarray
Camarray is a series of embedded stereo cameras and multiple camera solutions from Arducam. Upgraded from Arducam Sync Stereo Camera HAT, the Camarry can disguise up to 4 synched camera modules as a single camera slot connection to embedded systems like the Raspberry Pi, Jetson nano, and Xavier NX. With Arducam camarray, your camera connectivities are no longer limited to the camera connectors on the motherboard or the carrier board. Along with Arducam-provided camera drivers, more flexibility can be added to your multi-camera applications.
NOTE
1.The Camarray HAT does not support digital pan in half resolution combine and programmable pan speed control.
2.The Camarray HAT is not a crop and a half (halving the horizontal resolution of each camera and halving the field of view) but a compressed half (halving the horizontal resolution of each camera but leaving the field of view unchanged), so it does not need to support scan mode.
Common Specs
Image Sensor
| Sensor Model | OV9281 |
| Shutter Type | Global Shutter |
| Active Pixels | 1280*4×800 |
| Resolution | 1MP*4 |
| Image Sensor Format | Type 1/4″ |
| Pixel Size | 3μm×3μm |
| Color Filter Array | None(Monochrome) |
Lens Assembly
| Interchangeability | YES |
| F/NO | 2.8 |
| Focus Type | Manual Focus |
| Focusing Range | 30cm to infinity (when focused to infinity) |
| Effective Focal Length(EFL) | 2.8 mm |
| Field of View(FoV) | 70° Horizontal |
| Lens Mount | M12 Lens |
| IR Sensitivity | No IR filter (sensitive to IR light) |
Board
| HAT Size | 65×56 mm |
| Camera Board Size | 40×40 mm |
OV9281 Camera Board Mechanical Drawing
| Pin No. | Pin Name | Type | Description |
|---|---|---|---|
| 1 | 3V3 | Power | 3.3V power supply |
| 2 | FSIN | Input | Frame Sync Input |
| 3 | STB | Output | LED Strobe Output |
| 4 | GND | Ground | Ground |
| 5 | 3V3 | Power | 3.3V power supply |
| 6 | SDA | I/O | SCCB serial interface data I/O |
| 7 | SCL | Input | SCCB serial interface clock input |
| 8 | GND | Ground | Ground |
Connector & Cable
| Connector Interface | MIPI CSI-2 22-pin 2-lane |
| Ribbon Cable Length | 150mm (22-pin), 300mm(15-22pin), 73mm(15-22pin) |
| Pin # | Name | Description |
|---|---|---|
| 1 | GND | Ground |
| 2 | CAM_D0_N | MIPI Data Lane 0 Negative |
| 3 | CAM_D0_P | MIPI Data Lane 0 Positive |
| 4 | GND | Ground |
| 5 | CAM_D1_N | MIPI Data Lane 1 Negative |
| 6 | CAM_D1_P | MIPI Data Lane 1 Positive |
| 7 | GND | Ground |
| 8 | CAM_CK_N | MIPI Clock Lane Negative |
| 9 | CAM_CK_P | MIPI Clock Lane Positive |
| 10 | GND | Ground |
| 11 | CAM_IO0 | Power Enable |
| 12 | CAM_IO1 | LED Indicator |
| 13 | CAM_SCL | I2C SCL |
| 14 | CAM_SDA | I2C SDA |
| 15 | CAM_3V3 | 3.3V Power Input |
Driver Performance (With Official V4L2 Driver)
| Video Format | GERY(8-bit)/Y10P(10-bit) |
| Output Interface | 2-lane MIPI serial output |
| Output Formats | 8/10-bit BW RAW |
| Frame Rates | [email protected]*4 × 800(on RPi) [email protected]*4 × 800(on Jetson) |
Quick Start Guide
NOTE
Same operation on Xavier NX.The only difference is the driver package.
Hardware Setup
Check the Driver
This step is to check if you have other drivers installed. If you are sure that you don’t have other drivers installed, you can skip this step and go to install Arducam V4L2.
1.Driver type
There is three drivers for Jetson:
- Official IMX219 driver
- Arducam IMX477 driver
- Arducam V4L2 driver
NOTE
These three drivers are in conflict with each other and only one of the three options is available.
1) The Arducam IMX477 driver and Arducam V4L2 driver will automatically overwrite the other drivers and you don’t need to uninstall the other drivers.
2) If you want to use the official IMX219 driver, please type the following command to uninstall Arducam IMX477 driver or Arducam V4L2 driver.
sudo dpkg -r arducam-nvidia-l4t-kernel
3) Installing the Arducam IMX477 driver is the same process as installing the Arducam V4L2 driver, only the driver package (.deb) is different.
The product is driven by the Arducam V4L2.
2.Check your driver
Execute the following command and see if it responds. If there is no response, the corresponding driver is not installed.
#Check if you have installed Official IMX219 driver
dmesg | grep imx219
#Check if you have installed Arducam IMX477 driver
dmesg | grep imx477
#Check if you have installed Arducam V4L2 driver
dmesg | grep arducam
Install Arducam V4L2 Driver
Please go to this page for detailed instruction.
First Use
1.Install V4L2 python module
for Python3.x：
wget https://bootstrap.pypa.io/get-pip.py
After you type the following command and press【Enter】, it will take several minutes to finish downloading.
sudo python3 get-pip.py
sudo pip3 install v4l2
NOTE
The V4L2 of Python3.x has a known bug that requires a manual fix, and the following error occurs when you import the v4l2 module into Python3.x:
You can refer to this link to fix this bug:
https://bugs.launchpad.net/python-v4l2/+bug/1664158
2.Download the demo code
git clone https://github.com/ArduCAM/MIPI_Camera.git
3.Check whether the camera is detected
ls /dev/video0
4.Check the video format supported
v4l2-ctl --list-formats-ext
5.Display images in real time
If accessing Jetson via remote software (e.g., MobaXterm), the following command (executed only once) is required to display the image.
export DISPLAY=:0.0
Enter program directory:
cd MIPI_Camera/Jetson/Jetvariety/example/
Quadrascopic OV9281 camera currently supports the following commands:
#RAW8
python3 arducam_displayer.py -f GREY --width 5120 --height 800 -d 0
GREY for RAW8.
-w and -h indicate the width and height of the input image.
Press Ctrl+C to exit image display.
Dropped frame?
Select Arducam/arducam_displayer.py, right-click on “open with Text Editor“.
The circled value below represents the number of column in the display resolution. Try making the value lower if the frame is dropped.
6.Receive data without displaying the image
Dropped frames may exist due to platform performance. You can test the actual input frames by only receiving data without displaying the image.
#RAW8
v4l2-ctl --set-fmt-video=width=2560,height=800,pixelformat='GREY' --stream-mmap --stream-count=-1 -d /dev/video0
#RAW10
v4l2-ctl --set-fmt-video=width=2560,height=800,pixelformat='Y16 ' --stream-mmap --stream-count=-1 -d /dev/video0
7.Adjust exposure
Open two terminals, the first one is for executing the displaying images command, the second one is for executing the adjusting exposure command.
#Adjust exposure:
v4l2-ctl -c exposure=1000
#Check exposure parameters (minimum, maximum, default)
v4l2-ctl -l
For example, execute the below command in the second terminal:
v4l2-ctl -c exposure=4000
Turning up the exposure time results in a brighter image and a lower frame rate.
8. Adjust gain
Open two terminals, the first one is for executing the displaying images command, the second one is for executing the adjusting gain command.
#Adjust gain:
v4l2-ctl -c gain=12
#Check exposure parameters (minimum, maximum, default)
v4l2-ctl -l
For example, execute the below command in the second terminal:
v4l2-ctl -c gain=12
Turning up the gain results in a brighter image and no change in frame rate.
9.Change frame rate
Type the following command, change the value of “X” to change frame rate.
v4l2-ctl -c frame_rate=X
For example, you want to change the frame rate to 30. First, open two terminals, type the following command in the first terminal
python3 arducam_displayer.py -f GREY --width 5120 --height 800 -d 0 --fps
Type the following command in the second terminal to change frame rate:
v4l2-ctl -c frame_rate=30
Type the following command in the second terminal to view the frame rate range of the current display mode:
v4l2-ctl -l
Display the Image via VLC Media Player
Install VLC media player:
sudo apt-get install vlc
Open VLC media player:
1.Display the image
Press Ctrl+C
【Video device name】→ select “/dev/video0“→ click 【Play】.
It may be stuck due to Jetson performance and VLC media player.
2.Adjust exposure, gain, and frame rate
Press Ctrl+E, drag the sliders at “Exposure“, “Gain” and “Frame_rate” to make adjustments under the “v4l2 control” tab.
Display the Image via Mplayer
1.Install mplayer
sudo apt-get install mplayer
2.Display image
mplayer tv:// -tv driver=v4l2:device=/dev/video0
NOTE
Do not enter this command remotely, otherwise it will be stuck.
Or enter the following command to zoom the image. The image will appear full, but it will become stuck.
mplayer tv:// -tv driver=v4l2:device=/dev/video0 -zoom -x 1280 -y 200
“-x” “- y” denotes the width and height of the scaled image, and it is recommended to modify them in equal proportion.
External Trigger Mode
Hardware Setup
The external trigger mode has requirements on the connection method and electrical level of the external trigger signal.
The external trigger signal voltage should be 1.8V, and then send it to four cameras at the same time (It can be achieved through the breadboard ).
NOTE: 3.3V may burn out the device for the DOVDD of the camera is 1.8V.
The external trigger signal is connected to the FSIN test hole of the camera board (you may need to solder the row of pins by yourself. Note that GND should also be connected).
Software Setup
Method 1: Through terminal
Enter external trigger mode:
v4l2-ctl -c trigger_mode=1
Exit external trigger mode:
v4l2-ctl -c trigger_mode=0
Method 2: Through software interface
If you use VLC player, you can check “trigger_mode” under [v4l2 controls] tab to enter external trigger mode.
NOTE
After the external trigger signal is effective, the field signal will be effective after a while, which is mainly for the exposure time. If the external trigger frequency is very fast and the exposure time is long, because the field will not respond to the external trigger signal when it is effective, it will cause the display frame rate is lower than the external trigger frequency, then shorten the exposure time can improve the frame rate (but the image will become dim).
What’s Next
Here are the things you can do after this quick start:
- Check the Application Note for applications like using Arducam userland MIPI camera drivers.
- Join the discussion in our forum.
Application Note
Use with Arducam USB3.0 32b Camera Shield
| column | raw | FPS | bandwidths/MB |
| 5120 | 800 | 45.0 | 175.8 |
Please strictly follow the installation and operation order below, otherwise errors will occur!
1.Download File here.
2.Connect the camera module to the HAT then connect HAT to the USB Camera Shield in the case of a total power failure.
3.Power the HAT through micro USB port,then power the USB Camera Shield.
4.Run USBTest
5.Scan and select the camera device number
6.Select CfgFile
7.Click 【load】- Click【open】
8.Power down the HAT then power the HAT.
9.Click【play】
