SharpCap Forums

When I use my ASI224MC in Raw16 mode I see a nice debayered image on screen. The color looks good. When I capture a "snapshot" the .png appears monochrome at first glance. I can Debayer the image in PixInsight and see the color. What I am wonder is: does SharpCap capture 12-bit color (the max # of bits you can get from the ASI224MC)?

The reason I am asking is what I read when I look at the ZWO SDK rev 2.8 dated 2018-08-08.

The function ASIGetVideoData(int iCameraID, unsigned char* pBuffer, long lBuffSize, int iWaitms) says this about the buffer size:

• bufSize Byte length for RAW8 and Y8 bufSize >= image_width*image_height for RAW16 bufSize >= image_width*image_height *2 for RGB8 bufSiz >= image_width*image_height *3

So far RAW16 it says the buffer should have 2 bytes per pixel. Which sounds kind of monochrome. Actually, it says bufSize >= 2 bytes per pixel. Is the ZWO driver smart enough to give you full resolution color if you supply a buffer to handle 6 bytes per pixel?

I don't have my files in front of me, but the snapshot from SharpCap in RAW16 mode is IIRC size for 2 bytes per pixel.

Hi,

as you know, colour cameras have a colour Bayer matrix filter in front of the pixels so that each pixel only reads data for one of the three colours depending on the colour filter in front of it.

When you run the camera in raw mode, the data you collect is (almost) completely unprocessed – if the Bayer matrix on the camera is

RGRGRG
GBGBGB
RGRGRG
GBGBGB

Then the value of the top left pixel in the image is a red value, the pixel to the right of it is a green value and the first two pixels of the
our green and blue values respectively. These are stored in an image that looks monochrome – because each pixel only has one colour value – but when you zoom in on the image you will see the distinctive grid pattern caused by the different colours being represented at different pixel positions.

So, SharpCap fetches back the data from the ZWO camera at either one byte per pixel (RAW8) or two bytes per pixel (RAW16) and saves that data to file in exactly the same format. When you load the image file into a suitable image processing program such as pixinsight, the colour data is restored by applying a debayering algorithm. Up until the point where you do this, there is no need to store three separate colour channels for each pixel – doing so would only increase the size of the data by a factor of three without any gaining information.

Hope this helps, Robin