MEGA SPI Camera
General Introduction
Out of a 10-year iteration on our legacy SPI cameras, the brand-new Arducam Mega was released on Dec 20th, 2022 with multiple milestone improvements, targeting battery-powered IoT devices with embedded machine learning and artificial intelligence applications, making the development of computer vision applications seamless for edge devices.
High-level commands were built in the new SPI camera series so using high resolution cameras on MCUs becomes so much easier from then on. With the fully open source SDK that supports most platforms, Mega is the first SPI camera fitting any MCUs. These, together with super low power, quick-response and other enhancements, make Arducam Mega a real legendary game-changer in the industry.
Basic Facts
-
Portfolio
Mega Type
- Arducam Mega 3MP: SPI camera with 3MP Max. Resolution, Stock Lens
- Arducam Mega 5MP-AF: SPI camera with 5MP Max. Resolution, Auto-Focus Stock Lens
- Arducam Mega 3MP-M12: SPI camera with 3MP Max. Resolution, M12 Lens
- Arducam Mega 5MP-M12: SPI camera with 5MP Max. Resolution, M12 Lens
- Arducam Mega 3MP-NoIR: SPI camera with 3MP Max. Resolution, M12 Lens, W/O IR-Cut Filter
- Arducam Mega 5MP-NoIR: SPI camera with 5MP Max. Resolution, M12 Lens, W/O IR-Cut Filter
- For those who need more than a single camera, we also provide the solution to connect up to four Mega cameras to a single MCU via a multi-camera adapter board: The Pico4ML Pro. It is an evolution of the Pico4ML based on the Raspberry Pi RP2040 microcontroller and designed for TinyML applications.
-
SDK
The SDK is fully open-source and unprecedentedly supports all popular platforms: Raspberry Pi RP2040, Arduino 8-bit AVR, STM32, MSP430, ESP32, ESP8266 and more. C/C++ SDK and GUI tool for Windows with documentation are available.
-
Running Basic
Mega SPI cameras communicate with the host microcontroller over a four-wire SPI bus and draw a claimed 55-56mA while idle and 136-154mA while active. Image data from the rolling shutter sensors can be output in RGB, YUV, and JPEG formats, and both offer a 3.3mm focal length and a 68.75 degree field of vision.
Based on Arduchip*, an FPGA-based camera controller that handles complex and fast-speed camera video signal timing then sends the images to a large off-chip frame buffer.
The MCU only needs to send a capture command and read the images out in pretty slow byte-by-byte order whenever it has time to do it, and the image will not be lost since it is held still in the frame buffer until explicitly flushed.
This architecture tackles the issue of missing a dedicated camera interface with uniformed hardware and software structures.
Arduchip
*We put a brand-new, future-proof Arduchip in the camera, which not only gives the camera a huge performance boost, and the wakeup speed is also 10 times faster than before. And the new firmware comes with tons of optimizations to ensure compatibility, scalability, and developability.
-
Application
Potential applications include assets monitoring, wildfire monitoring, environmental observation, remote meter reading, TinyML applications, and so on.
Features
| Easy Using: The built-in high-level commands allow users to manipulate the camera through API access, like using a DSLR camera with button clicks. Also refer to demonstrations |
| Low Pin Count: Different from its predecessors, we removed two I2C interfaces and now only 6 pin left. 4 for SPI, 2 for power. |
| Extreme Compatibility: It fits any microcontroller that has a single standard SPI interface (either native or mimic one). The HAL (hardware abstraction layer) design is used so that the adaption to different platforms becomes much simpler. |
| High Resolution: Max resolution is up to 2048 x 1536 for Mega 3MP and 2592 x 1944 for Mega 5MP. |
| Impressive Image Quality: Please refer to sample photos & performance test |
| Quick Response: Wake-up is as quick as 94ms for Mega 5MP and 42ms for Mega 3MP. Because the sensor is now directly talking to the chip. |
| Power Efficiency: The power consumption is lowered down to under 200 mW for idle mode, and under 650mW for working mode. |
| Multiplexing: You can port 4 Mega onto the adapter (The Pico4ML Pro) and the 4 quads capture images simultaneously then read out sequentially. |
More Resources
You can refer to the following link to discover more details:
Arducam Mega SPI Camera
Sample Photos
 |
 |
 |
Performance Test
 |
 |
 |
 |
Demonstrations
Common Specifications
| Specifications | MEGA-3MP | MEGA-5MP |
|---|---|---|
| Interface | SPI (4-wire) | SPI (4-wire) |
| Speed | 8Mhz | 8Mhz |
| Optical Size | 1/4″ | 1/4″ |
| Shutter Type | Rolling Shutter | Rolling Shutter |
| Max Resolution | 2048×1536 | 2592x1944 |
| Still Resolution | 320×240 640×480, 1280×720 1600×1200, 1920×1080 2048×1536 |
320×240 640×480, 1280×720 1600×1200, 1920×1080 2592×1944 |
| Camera Case | ABS Case | ABS Case |
| Power Supply | 3.3V/5V | 3.3V/5V |
| Electric Current | 56mA-136mA | 55mA-154mA |
| Power Consumption | 185mW-585mW | 182mW-650mW |
| Wake-Up Time | 42ms | 94ms |
| Output Format | RGB/YUV/JPEG | RGB/YUV/JPEG |
Product
| Product Image | SKU | Pixels | Max Resolution | Lens Type | Focus Type | Field of View(HxV) | IR Sensitivity | Wake-Up Time | Power Consumption |
|---|---|---|---|---|---|---|---|---|---|
 |
B0400 | 3MP | 2048×1536 | Stock Lens | Fixed Focus | 68.75 degree (Diagonal) |
Integral IR-cut filter | 42ms | 185mW-585mW |
 |
B0434 | M12 Lens | Manual Focus | 90°x75°x56.4° | Integral IR-cut filter | ||||
 |
B0435 | without IR-cut filter | |||||||
 |
B0401 | 5MP | 2592×1944 | Stock Lens | Autofocus | 68.75 degree (Diagonal) |
Integral IR-cut filter | 94ms | 182mW-650mW |
 |
B0436 | M12 Lens | Manual Focus | 90°x75°x56.4° | Integral IR-cut filter | ||||
 |
B0437 | without IR-cut filter |
Related Docs
Quick Start Guide for Arducam Mega Camera
Additional Resources
For more details about the timing description, Register information, etc., please refer to the following datasheet:
Arducam MEGA SPI Camera - Datasheet