In an earlier post, I shared plans for a CYE32 vision sensor based around the AT32UC3B1256 microcontroller. After working with this vision sensor for a couple months, it has become a useful platform for trying out new vision chips and new image processing algorithms.
I put together a short video that shows off some of the features of the CYE32, including variable downsampling and various optical flow algorithms. The vision chip used in the video is the Faraya 64 (with 64×64 pixel array), but the CYE32 supports the Firefly and the new Stonyman class chips as well.
The CYE32 outputs bytes via I2C. These bytes are relayed by a custom I2C-to-USB board to the computer. Commands can be sent to the sensor to change image parameters in real-time, and various datasets can be read from the sensor and displayed in the output window.
I’d be interested to know whether there is a market for a standalone open-source sensor such as this. With the addition of a JTAG programmer, a user would be able to select any part (or parts) of an image, process that image within the limits of the microcontroller, and then produce an I2C output.
Take a look at the video and leave your feedback below.
You must log in to post a comment.