Product Specifications🔗

The PlanktoScope OS includes all software which needs to run on the PlanktoScope's hardware to provide the overall functionality of a PlanktoScope. Product specifications for the PlanktoScope OS are listed below for ranges of software version numbers. To see software versions listed individually in chronological order, refer to the project release notes or the software changelog. To understand how to interpret software version numbers, refer to our description of the PlanktoScope OS's version numbering system.

v2023.9.0 - v2024.0.0🔗

Functionalities🔗

Regular operation:

Image acquisition: stop-flow imaging, JPEG image output
On-board image processing: detection and segmentation of objects (batch-processing only)
User interfacing: graphical interface accessible through web browser of a connected phone, tablet, or computer
Export of data for uploading to EcoTaxa

Advanced operations:

User interfacing: web browser interfaces for system administration, system monitoring, and troubleshooting
Automation: MQTT-based API
Application deployment: ability to add software as OCI containers using Docker

Base operating system🔗

Distro: Raspberry Pi OS 11 (bullseye)
Binary target architecture: 32-bit only (armhf, also known as armv7)

Supported hardware🔗

Minimum for image acquisition (but not sufficient for on-board image processing):

PlanktoScope: hardware v2.1 with Raspberry Pi 4 Model B computer
Memory: 1 GB RAM
Storage: 8 GB capacity

Minimum for full functionality, including on-board image processing:

Memory: 4 GB RAM

Recommended:

PlanktoScope: hardware v2.5 or v2.6 with Raspberry Pi 4 Model B computer
Storage: 32 GB capacity

Forwards-incompatibilities:

These software versions are unable to run on the Raspberry Pi 5 computer.

Backwards-incompatibilities:

These software versions might still work on a Raspberry Pi 3 Model B+ computer or a Raspberry Pi 4 Model B computer with 1 GB of RAM, but compatibility is not tested.

System performance🔗

With minimum supported hardware for full functionality:

On-board image processing: a dataset of 400 raw images is processed in approximately 1.5 to 2 hours

v2.3🔗