• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 99 subscribers
  • Views 344 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DM6446 - How to direct CCDC output to DDR2 in Raw image capture mode

Arun12185
Intellectual 390 points

Hello,

My objective is to capture images from a VGA sensor connected to VPFE of DM6446 (raw mode). I am using dm6446 EVM and the sensor gives 10-bit parallel data along with HD, VD, PCLK.

I have taken mt9t001.c example code and I am in the process of making code changes in davinci_vpfe.c, ccdc_davinci.c according to our sensor.

Also I have taken prev_vga_example.c code as an application code to capture images (from the link http://www-s.ti.com/sc/techlit/sprc434.gz).

Initially I want to store raw images into the DDR2 and access it from the application layer to save them in bitmap/tiff format. How can I bypass previewer and other modules and store image data into DDR2 directly from CCDC output and also access stored images from user space? Please provide any available links to any example code/application notes which explains the same or suggest any guidelines to achieve this?

MV and dvsdk versions are mv_pro_4.0.1, dvsdk_1_30_01_41 respectively.

Thanks for your help

--

Best Regards

S Arun

  • 0
    TI__Mastermind 37340 points

    Arun,

    By default, our CCD driver captures data to DDR2; from there it can be fed to preview engine (or not), stored in a file, read by an application...  I would recommend looking at the examples included in the DVSDK as opposed to the ones in the App notes, these App Notes are quite old and the software version they were based on are likely incompatible with newer DVSDKs.

  • 0 in reply to Juan Gonzales
    Expert 1225 points

    HI.

    Can you tell me more specifically abt those examples ?? that will be gr8 help !! thank you

  • 0 in reply to Digant Desai
    TI__Mastermind 37340 points

    I am particularly referring to the DVSDK demos (encode, decode, encodedecode); in particular, the encode demo takes video data in and copies it to DDR2 (you have access to this data in the user space encode app), from there it can pass this DDR2 buffer to be encoded before finally writting the encoded data to a file. 

    Starting from this demo, you can take out the steps to encode, and may need to modify driver to get rid of preview color space conversion, since this step is not accessible at the user space level.

    Also, I would recommend move towards a newer DVSDK were we have implemented better more modular driver support along with a DMAI laibrary layer that has simpler examples then the DVSDK you cited above.