OpenDataCam 2.0 – An open source tool to quantify the world
2019 Sep
OpenDataCam is an open source tool to quantify the world. It quantifies and tracks moving objects with live video analysis. It is designed to be an accessible, affordable and open-source solution to better understand interactions in urban environments.
OpenDataCam never records any photo or video data. The system only saves surveyed meta-data, in particular the path an object moved or number of counted objects at a certain point. The novelty of OpenDataCam is, that everything happens on location, while no visual data is saved or sent to online cloud processing.
OpenDataCam runs on Linux and CUDA GPU enabled hardware. It is optimized for the NVIDIA Jetson Board series. The most affordable setup runs on a Jetson Nano (low cost, credit-card sized GPU-computer) combined with other other off-the-shelf equipment (webcam, power supply, housing), this entire setup is priced around $150. All software is based on open source components and runs completely locally. The software features a friendly user interface and is currently optimised for detecting and counting traffic participants, but is not limited to that.
Both software and hardware setup are documented and offered as an open source project, to underline transparency and full disclosure on privacy questions. The simple OpenDataCam setup allows everybody to become an urban data miner.
Features
How can OpenDataCam see? – OpenDataCam consists of a camera attached to a mini computer. It’s running an object detection algorithm that counts and tracks moving objects. The board is equipped with a GPU, which allows it to process many parallel image threads at once. The perfect tool for image analysis and video processing.
Yolo – On the software side OpenDataCam is running YOLO, an object detection library. YOLO is trained to detect objects in images. The attached camera feeds YOLO with a video, YOLO then outputs all objects in each frame.
Housing – OpenDataCam is not tied to a specific housing, instead you can design your own or use an already existing casing. We used the Wildlife Cam Casing by Naturbytes to protect the hardware in the wild. Have a look at this tutorial to see how the set up the the Jetson NANO in the Wildlife Cam Casing.
You own the data! OpenDataCam is not connected to the cloud, you’re in full control of when to gather it. Via the interface you can specify which areas of the picture objects should be counted. Also the camera tracks how things move through the frames. Via an easy export function you can access the tally and traces and use it in any thinkable way. Going all-in? Use the API to get access to the raw data points.
Ethics
Transparency – The discussion around the use of artificial intelligence in public space is ongoing. We believe OpenDataCam is a contribution to this discussion, as it offers an offline, transparent, thus privacy respecting alternative. Paramount to the open source access is the responsible use of this technology we request from every user.
Privacy – As OpenDataCam is an experimental prototype with public facing applications, privacy is something we tried to bake into the design of the tool. After getting in touch with Sensor Labs privacy label, we’re currently finding ways to apply the label to OpenDataCam.
