Hello,
We plan on interfacing via SPI to our PIXY’s. We’d like to hit the ground running when we get ours (February?).
Is there a user manual / format definition for the SPI interface? If not, can the source as it stands now be posted so we can begin to code to that interface?
Thanks!
Hi Matthew,
An initial SPI message format was given on the “Kickstarter”:http://www.kickstarter.com/projects/254449872/pixy-cmucam5-a-fast-easy-to-use-vision-sensor page (under “Communication”). It has changed slightly to this:
0: start word: 0xaa55
1: checksum (fields 2 thru 6 added up)
2: model (signature number)
3: x center
4: y center
5: width
6: height
7: end word: 0xccaa
Each field is 16 bits. I suspect it will change again, but not by much. We’ll be sure to post the final format once the devices ship.
You can also see the code that handles the SPI messages at our “Github repo”:https://github.com/charmedlabs/pixy/blob/master/device/video/blobs.cpp in the getBlock() method.
Let me know if you have any more questions.
Scott
Thanks Scott, that makes sense.
Any provision for doing an image dump out the SPI port? (i.e. a snapshot of the last frame for example)?
Thanks again,
Matt
Hi Matt,
Not yet. This could easily be added later, but it would be slow. Unless you’re limited to SPI, USB would be a much better bet. I doubt this will be included at launch, but we (or someone else) could always add it later if the demand is there.
Scott
Hi Scott,
One year after Matt’s question, is there today any plan to add a raw image snapshot dump through the serial USART?
Thanks in advance for your answer
Xavier
Hello Xavier,
To answer your question, we don’t have a plans for this feature. It’s a natural feature I suppose, but Pixy’s main use is to process the images onboard and only provide the results, so your micro doesn’t get bogged down. We’re always interested in hearing how someone might use this feature though. Please describe your application. And what kind of micro do you plan on using?
Edward