JeVois is an open-source machine vision camera for Makers. The platform consists of a video sensor, a quad-core CPU, USB video, and a serial port, all in a tiny package.
Insert a microSD card loaded with the provided open-source machine vision algorithms (including OpenCV 3.1 and others), connect to your desktop, laptop and/or Arduino, and give your DIY projects the sense of sight immediately.
As its Kickstarter page explains, the JeVois framework operates as follows: video is captured from the camera sensor, processed on the fly through algorithm directly on the camera’s own processor, and the results are streamed over USB to a host computer and/or over serial to an MCU.
To the host computer, the JeVois smart camera is simply just another USB camera. Different vision algorithms are selected by changing USB camera resolution and framerate. Users or machines can also interact with JeVois, change its settings, or listen for text-based vision outputs over serial link.
JeVois is currently programmed with three modes. First, a development mode that outputs a demo display over USB and communicates results. Next, a text-only mode that gives strings of text from the smart camera without USB output. Last but not least, a pre-processing mode that outputs video intended for machine consumption which can then be further processed by the host computer.
The quad-core ARM-V7 microprocessor runs at 1.34GHz and boasts 256 MB DDR3–1600 RAM and a dual-core Mali-400 GPU. The 1.3MP camera can operate at speeds up to 120fps. What’s more, the JeVois smart camera can work as a standalone Linux computer without a host PC.
You can learn more about the open-source camera, including its specs and requirements, on Kickstarter.