Quantization & Dithering - 2023.2 English

Vitis Libraries

Release Date
2023-12-20
Version
2023.2 English

This algorithm dithers input image using Floyd-Steinberg dithering method. It is commonly used by image manipulation software, for example when an image is converted into GIF format each pixel intensity value is quantized to 8 bit i.e. 256 colors.

API Syntax

template <int IN_TYPE, int OUT_TYPE, int ROWS, int COLS, int SCALE_FACTOR, int MAX_REPRESENTED_VALUE, int NPC, int XFCVDEPTH_IN = _XFCVDEPTH_DEFAULT, int XFCVDEPTH_OUT = _XFCVDEPTH_DEFAULT>
void xf_QuatizationDithering(xf::cv::Mat<IN_TYPE, ROWS, COLS, NPC, XFCVDEPTH_IN>& stream_in,
                             xf::cv::Mat<OUT_TYPE, ROWS, COLS, NPC, XFCVDEPTH_OUT>& stream_out)

Parameter Descriptions

The following table describes the template and the function parameters.

Table 685 Table Quantization & Dithering Parameter Description
Parameter Description
IN_TYPE Input pixel type. It should be XF_8UC1, XF_8UC3, XF_10UC1, XF_10UC3, XF_12UC1, XF_12UC3, XF_16UC1, or XF_16UC3. Note XF_<PIXEL_BITWIDTH>UC<NUM_CHANNELS>
OUT_TYPE Output pixel type. It should be XF_8UC1, XF_8UC3, XF_10UC1, XF_10UC3, XF_12UC1, XF_12UC3, XF_16UC1, or XF_16UC3. Output PIXEL_WIDTH should less than or equal to input PIXEL_BITWIDTH
ROWS Maximum height of input and output image
COLS Maximum width of input and output image. In case of N-pixel parallelism, width should be multiple of N.
SCALE_FACTOR The SCALE_FACTOR must be power of 2 & less than or equal to 2^(output PIXEL_BITWIDTH)
MAX_REPRESENTED_VALUE The MAX_REPRESENTED_VALUE must be equal to equal to 2^(input PIXEL_BITWIDTH)
NPC Number of pixels to be processed per cycle; possible options is XF_NPPC1 or XF_NPPC2
XFCVDEPTH_IN Depth of the input image.
XFCVDEPTH_OUT Depth of the output image.
stream_in Input image
stream_out Output image

Resource Utilization

The following table summarizes the resource utilization of the kernel in different configurations, generated using Vitis HLS 2020.2 tool for the Xilinx xcu200-fsgd2104-2-e FPGA, to process a RGB image with a resolution of 1024x676 & pixel width 16 bit and quantized it to 8 bit pixel width.

Table 686 Table Quantization & Dithering Resource Utilization Summary
Operating Mode Operating Frequency (MHz) Utilization Estimate
BRAM_18K DSP_48Es FF LUT CLB
1 pixel 300 11 10 7592 5765 1582
2 pixel 300 14 12 8150 6945 1749

Performance Estimate

The following table summarizes a performance estimate of the kernel in different configurations, as generated using Vitis HLS 2020.2 tool for the Xilinx xcu200-fsgd2104-2-e FPGA, to process a RGB image with a resolution of 1024x676 & pixel width 16 bit and quantized it to 8 bit pixel width.

Table 687 Table Quantization & Dithering Function Performance Estimate Summary
Operating Mode

Operating Frequency

(MHz)

Latency Estimate

Max (ms)

1 pixel 300 2.8
2 pixel 300 1.58