Skip to content

64MP OwlSight

Introduction

Arducam has released the highly influential and industry benchmark Arducam 64MP Hawkeye autofocus embedded camera for Raspberry Pi platform in 2022. Since the product was released, it has been widely praised. However, the camera, which is completely based on Arducam's independent research and development, cannot be natively supported by the Raspberry Pi platform. Users need to download the corresponding driver installation package to drive and use the camera.

Now, Arducam is coming with the latest Arducam 64MP OwlSight OV64A40 autofocus camera module. The biggest feature of the new 64MP OV64A40 OwlSight camera is that it will be fully supported by the mainstream software: libcamera apps.

In addition, we have created a complete ecosystem for it, with accessories, products, and solutions available, which will greatly save you the energy and time spent in developing and implementing projects. We also support a deeper freedom in customization. You can share any of your ideas with us, and we will use our expertise and services to help you realize any project.

Key Features

  • Native Libcamera Support: Directly compatible with Raspberry Pi's mainstream libcamera software, eliminating the need for custom driver installation.
  • Ultra-High Resolution: Captures stunningly clear images with a maximum resolution of 9248 × 6944 (64MP).
  • Excellent Photosensitive Performance: Utilizes a large 1/1.32″ optical format and class-leading pixel size for superior light gathering.
  • High Frame Rates: Achieves high frame rates even at high resolutions.
  • Dual Mode Focus: Supports both CDAF (Contrast Detection Autofocus) and software-controlled manual focus.
  • Rich Ecosystem: A full range of accessories and solutions are available to accelerate project implementation.

Specifications

Image Sensor

Parameter Specification
Image Sensor OV64A40
Still Resolution 64 Megapixels
Color Filter Type Color (Quad-Bayer)
Shutter Type Rolling Shutter
Optical Size 1/1.32″
Pixel Size 1.008 µm × 1.008 µm
Sensor Active Area 9248 (H) × 6944 (V)
Output Formats JPEG / YUV / RGB / RAW10
IR Sensitivity Integral IR-cut Filter, visible light only

Lens

Parameter Specification
Focus Type Manual / Autofocus
Focal Length 6.65mm
Aperture (F.NO) F1.9
Field of View (FOV) 84°(D) × 68°(H) × 56°(V)
Default Focus Range 12cm ~ ∞

Electrical & Mechanical

Parameter Specification
Power Requirements 3.3V / 300mA
Camera Board Size 25 × 24 mm

Frame Rate & Resolution

The OV64A40 camera's frame rate can be adjusted by setting the MIPI link frequency. This is controlled via the link-frequency parameter within the dtoverlay line in your config.txt file.

Two link frequencies are supported:

High-Speed Mode (link-frequency=456000000)

Resolution Frame Rate
9152 × 6944 2.6fps
8000 × 6000 2.5fps
4624 × 3472 10fps
3840 × 2160 20fps
2312 × 1736 30fps
1920 × 1080 60fps

Low-Speed Mode (link-frequency=360000000)

Resolution Frame Rate
9152 × 6944 2fps
8000 × 6000 2.5fps
4624 × 3472 7.6fps
3840 × 2160 14.8fps
2312 × 1736 26.7fps
1920 × 1080 45fps

Product Catalog

Product Image SKU Resolution Sensor Interface Features Focus Type Focus Distance Field of View(DxHxV) IR Sensitivity
B0483 64MP OV64A40 MIPI CSI-2 Ultra High Resolution Manual/Autofocus 12cm-infinite 84°(D)x68°(H)x56°(V) Integral IR-cut Filter

Platform Compatibility

Supported Operating Systems

Platform Trixie Bookworm (rpicam)
Raspberry Pi 5 / CM5 ✅ ✅
Raspberry Pi 4B / 3B+ / 3A+ / Zero / Zero 2 W ✅ ✅
Raspberry Pi CM3 / CM3+ / CM4
Requires adapter board		 ✅ ✅

Operating System Support

This documentation focuses on Trixie, Bookworm to align with the official Raspberry Pi OS release cycle.

Hardware Connection

Connect your camera module by selecting the tab that corresponds to your Raspberry Pi model.

Connection Principle

The core principle is to connect the camera to the MIPI CSI (Camera Serial Interface) port on your Raspberry Pi. Ensure the cable is inserted correctly with the metal contacts properly seated.

Using the 15-pin to 22-pin FPC cable, connect your camera to one of the dual MIPI CSI ports on your Raspberry Pi 5. Ensure the black tab on the cable faces the HDMI ports.

Hardware connection diagram for Raspberry Pi 5
Figure 1: Connecting the camera to a Raspberry Pi 5

Using the 15-pin to 15-pin FFC cable, connect the flexible flat cable to the full-size MIPI CSI port located between the audio/video jack and the HDMI port. The blue tab on the cable should face the Ethernet port.

Hardware connection diagram for Raspberry Pi 4B
Figure 2: Connecting the camera to a Raspberry Pi 4B

If you cannot find your Raspberry Pi model above, please refer to the following doc for common hardware connection method of RPI Camera:

Quick Start Guide - Hardware Connection

Software Configuration

This section guides you through the necessary software configuration to enable your OwlEye camera, followed by instructions on how to use it. The process is divided into these main parts:

  1. System Configuration: The required first step for the Raspberry Pi to recognize the camera. A reboot is required for changes to take effect.
  2. Camera Usage: Basic commands to preview and capture images.
  3. Advanced Usage: Optional steps for controlling features like autofocus.
  4. Long Exposure Photography: Instructions for capturing long exposure images.

1. System Configuration

Note

The Arducam OwlEye camera is natively supported on Raspberry Pi OS Bookworm and Trixie only. Bullseye and older versions are not supported.

The configuration process is similar for all compatible Raspberry Pi models, with the configuration file always located at /boot/firmware/config.txt.

Raspberry Pi 5 & CM5

  1. Open the configuration file for editing: 

    sudo nano /boot/firmware/config.txt

  2. Find the line camera_auto_detect=1 and change it to 0.

  3. Locate the [all] section and add one of the following dtoverlay lines below it. This enables the camera and sets its speed.

    This mode offers maximum stability, especially with longer camera cables. 

    dtoverlay=ov64a40,link-frequency=360000000

    This mode provides higher frame rates. 

    dtoverlay=ov64a40,link-frequency=456000000

  4. Save the file and reboot your Raspberry Pi. 

    sudo reboot

Using the CAM0 Port

The Pi 5 defaults to CAM1. If your camera is connected to the port labeled CAM0, you must append ,cam0 to the dtoverlay line. For example: dtoverlay=ov64a40,cam0,link-frequency=360000000

Raspberry Pi 4, Pi Zero to Pi 3, & CM3/4

The process is the same for these models, as they all run Bookworm/Trixie with the same file structure.

  1. Open the configuration file: 

    sudo nano /boot/firmware/config.txt

  2. Set camera_auto_detect=0.

  3. Add the dtoverlay line under [all], choosing your desired speed mode.

    • For Low-Speed: dtoverlay=ov64a40,link-frequency=360000000
    • For High-Speed: dtoverlay=ov64a40,link-frequency=456000000

  4. Save the file and reboot.

Using CAM0 on Compute Modules

If you are using a Compute Module (CM3/4) and have the camera connected to the CAM0 interface, append ,cam0 to your chosen dtoverlay line.

2. Camera Usage

After rebooting, your camera is ready. On Bookworm and Trixie, the rpicam-apps suite is used.

  1. Confirm camera detection by listing available cameras: 

    rpicam-still --list-cameras

  2. Start a live preview (Press Ctrl+C to exit): 

    rpicam-still -t 0

  3. Capture an image named test.jpg after a 5-second preview: 

    rpicam-still -t 5000 -o test.jpg

3. Advanced Usage: Focus & Zoom Control

The OwlEye camera supports several focus modes.

  • Continuous Autofocus: 

    rpicam-still -t 0 --autofocus-mode continuous

  • Single Autofocus (before preview starts): 

    rpicam-still -t 0 --autofocus-mode auto

  • Manual Focus with Keyboard Control: For interactive manual focus, Arducam's extended libcamera-apps are recommended as they provide keyboard controls.

    1. Install Arducam's libcamera software: 
      wget -O install_pivariety_pkgs.sh https://github.com/ArduCAM/Arducam-Pivariety-V4L2-Driver/releases/download/install_script/install_pivariety_pkgs.sh
chmod +x install_pivariety_pkgs.sh
./install_pivariety_pkgs.sh -p libcamera_dev
./install_pivariety_pkgs.sh -p libcamera_apps
    2. Run in manual focus preview mode: 
      rpicam-still -t 0 --autofocus-mode manual -k
    3. Use the following keys in the terminal window to adjust focus: F (trigger auto), A (move lens away), D (move lens towards).

  • Zoom Control: Please refer to the following program to control the zoom function:

4. Long Exposure Photography

The OwlEye camera is capable of long exposure photography, making it suitable for low-light conditions. The maximum possible exposure time depends on the selected resolution and the MIPI link frequency.

Maximum Exposure Time

Low-Speed Mode (link-frequency=360000000)

Resolution Max Exposure Time (seconds)
1280 × 720 204.01
1920 × 1080 304.22
2312 × 1736 336.43
3840 × 2160 492.57
4624 × 3472 608.45
8000 × 6000 1059.42
9248 × 6944 1173.95

High-Speed Mode (link-frequency=456000000)

Resolution Max Exposure Time (seconds)
1280 × 720 146.74
1920 × 1080 196.85
2312 × 1736 279.17
3840 × 2160 378.04
4624 × 3472 472.44
8000 × 6000 860.78
9248 × 6944 910.88

Capturing Long Exposure Images

To perform a long exposure shot, you need to set the --shutter parameter, which is specified in microseconds (μs).

  • --shutter: Sets a fixed exposure time in microseconds. (1 second = 1,000,000 μs)
  • -t: Sets the total program run time in milliseconds (ms). This must be longer than the shutter speed.
  • -o: Specifies the output filename.

Usage Examples

Set exposure time to 1 second: 

rpicam-still -t 2000 --shutter 1000000 -o shutter_1s.jpg

Set exposure time to 5 seconds: 

rpicam-still -t 6000 --shutter 5000000 -o shutter_5s.jpg

Set exposure time to 10 seconds: 

rpicam-still -t 11000 --shutter 10000000 -o shutter_10s.jpg

Comparison

The following image demonstrates the dramatic difference long exposure can make in low-light environments.

Long exposure comparison showcasing enhanced low-light visibility
Low-light image comparison: exposure time from 1s to 10s

Additional Resources