Hi, Thanks for the replys guys (useful sites, Franz). What I'm trying to do is figure out how YUYV (YCbCr 4:2:2) is encoded in the picture as an array of unsigned chars. My aim is to "crop" the image just after it is captured (as VIDEO_PALETTE_YUYV). I have already computed the coordinates necessary: from pixel (x,y)=(72,56) crop a rectangle of (width,height)=(660,382). These coordinates correspond to a rectangle in the middle of the image that I am capturing (PAL mode, 768x576), outside of which data can be safely discarded as useless. I got these coordinates by saving an image captured in RGB24 as a file on disk, and then with GIMP I found the coordinates pf those points of the rectangle that I am interested in (see above). The challenge now is to actually use these coordinates on the YUYV input directly; that is once the full frame is captured, I want to make a cropped copy of it according to the coordinates. Hence, for all subsequent copies of the image (and also YUYV to RGB24 colourspace transformation for display purposes) I will save time by transferring a smaller image than the original full frame. Can anyone help, please? Thanks. /G
