Convert interleaved CbCr frame to separate components
Computer Vision System Toolbox Support Package for Xilinx Zynq-Based Hardware
The Deinterleave block accepts an input video frame consisting of interleaved Cb and Cr values. The output is separate Cb and Cr frames. Use this block to convert an interleaved frame to separate Cb and Cr frame components for connecting to Computer Vision Toolbox™ blocks. The frame size of the Cb and Cr output components is half the size of the associated Y component.
The FPGA data path for YCbCr 4:2:2 format consists of a Y component and a time-interleaved CbCr component. The pixel-streaming subsystem targeted for the user logic on the FPGA must have these two input and output ports. When you convert the pixel stream to frames, use dummy data (duplicate the CbCr component) for the third component. The resulting frame has 3 full-sized components. To connect to Computer Vision Toolbox blocks, drop the duplicate component and separate the interleaved frame into the half-size Cb and Cr component frames. See the Pixels to Frame for YCbCr 4:2:2 block in the YCbCr 4:2:2 HDL Pixel-Streaming Algorithm template, and the example models.
CbCr— Interleaved video frame
Interleaved video frame, specified as a matrix. The matrix represents the chrominance component of an image in YCbCr 4:2:2 format. It contains interleaved Cb and Cr pixel values.
Cb— Blue chrominance frame
Blue chrominance frame, returned as a matrix. The matrix is half the size of the input matrix.
Cr— Red chrominance frame
Red chrominance frame, returned as a matrix. The matrix is half the size of the input matrix.