Monday, September 16, 2013

OpenMV Camera Module

This a quick update on my camera project (OpenMV)... I've finally received the long overdue PCBs today and assembled one:


I haven't written any code yet, but I've managed to get it into DFU mode and upload a blinky, it seems to be working fine so far, no smoke :) I will post another update or a video as soon as I get something cool running.


15 comments:

  1. Hi,

    I am from Bangalore, India. I am looking to build a very tiny device that has camera + Bluetooth capabilities. The processing of the image itself will happen on an android phone connected via Bluetooth.

    It is not mandatory to have JPEG encoding in the camera. I can just transfer the raw stream to the connected phone via Bluetooth. Will the OpenMV camera module work for me? Or should I go with the nRFcam (but that has Wi-Fi and not Bluetooth I suppose)?

    Can you suggest the best combination of MCU + Camera + Bluetooth radio?

    ReplyDelete
    Replies
    1. I'm still working on this module, I don't know yet if it works, I'm getting an image, but it's all messed up.. you could look into the newer Nordic BLE chips and TCM82xx sensors..

      Delete
    2. Thanks! Which MCU would you recommend? ARM Cortex M0 or M3 or Arduino?

      Delete
    3. I would use anything with a DCMI it makes life easier..

      Delete
  2. Replies
    1. yes, I will share the schematics and source code soon, I'm still working on it.

      Delete
  3. This is awesome. I am doing some projects that need to look for colored objects (like a red laser dot) and it looks like this is just the thing I was looking for. I am very very interested in it and would be very interested in testing it... or simply buying one if you ever decide to sell them (please?)

    ReplyDelete
  4. Replies
    1. https://github.com/iabdalkader/openmv

      Delete
  5. Is there any kind of manual long exposure available?. I'm thinking in possible astronomical purposes of this module, like an 'automatic guider' for astrophotography.

    All it would need to do is to 'detect' a group of pixels that are brighter (a star) and follow them if they move. But for this to work, I'm afraid that long exposures (of 0.5-2 seconds) would be needed to increase the sensitivity.

    ReplyDelete
    Replies
    1. yes, exposure and some other functions can be set manually, not sure how long though.

      Delete
  6. Hey,

    I am currently working on interfacing a OV2640 with a STM32F4-Discovery board. I found your source codes and schematics very helpful in this regard.

    But I have encountered an error which I haven't been able to solve. I am trying to send JPEG images over UART. I have used the driver and regs file that you have provided with your code. But, I am getting a overflow signal after only a few read operations after start of frame.

    I get two VSYNC signals followed a few HSYNC signals and an overflow signal. Thus the DMA transfer gets aborted. Could you help me with this?

    ReplyDelete
    Replies
    1. you shouldn't get HSYNCs with JPEG (there are no lines in JPEG) and you should only get one VSYNC per frame, there's another post about reading JPEG you might find it useful.

      Delete
  7. Thanks for replying so quickly. According to the STMF4 reference manual though, the DCMI interface uses HSYNC as a data valid signal. So shouldn't I get several packets of data between 2 VSYNC interrupts? I read through the JPEG post as well. I am using the same method as well for aborting the DMA incase of JPEG.

    I actually printed the interrupt calls, this is what I got,

    Line interrupt (Hsync), vsync, vsync, line, vsync, vsync after which the code goes to the end of the program. Could you guess at the problem?

    ReplyDelete