Video and Vision Processing Suite Intel® FPGA IP User Guide

ID 683329
Date 6/26/2023
Public

A newer version of this document is available. Customers should click here to go to the newest version.

Document Table of Contents
1. About the Video and Vision Processing Suite 2. Getting Started with the Video and Vision Processing IPs 3. Video and Vision Processing IPs Functional Description 4. Video and Vision Processing IP Interfaces 5. Video and Vision Processing IP Registers 6. Video and Vision Processing IPs Software Programming Model 7. Protocol Converter Intel® FPGA IP 8. 3D LUT Intel® FPGA IP 9. AXI-Stream Broadcaster Intel® FPGA IP 10. Bits per Color Sample Adapter Intel FPGA IP 11. Chroma Key Intel® FPGA IP 12. Chroma Resampler Intel® FPGA IP 13. Clipper Intel® FPGA IP 14. Clocked Video Input Intel® FPGA IP 15. Clocked Video to Full-Raster Converter Intel® FPGA IP 16. Clocked Video Output Intel® FPGA IP 17. Color Space Converter Intel® FPGA IP 18. Deinterlacer Intel® FPGA IP 19. FIR Filter Intel® FPGA IP 20. Frame Cleaner Intel® FPGA IP 21. Full-Raster to Clocked Video Converter Intel® FPGA IP 22. Full-Raster to Streaming Converter Intel® FPGA IP 23. Genlock Controller Intel® FPGA IP 24. Generic Crosspoint Intel® FPGA IP 25. Genlock Signal Router Intel® FPGA IP 26. Guard Bands Intel® FPGA IP 27. Interlacer Intel® FPGA IP 28. Mixer Intel® FPGA IP 29. Pixels in Parallel Converter Intel® FPGA IP 30. Scaler Intel® FPGA IP 31. Stream Cleaner Intel® FPGA IP 32. Switch Intel® FPGA IP 33. Tone Mapping Operator Intel® FPGA IP 34. Test Pattern Generator Intel® FPGA IP 35. Video and Vision Monitor Intel FPGA IP 36. Video Frame Buffer Intel® FPGA IP 37. Video Frame Reader Intel FPGA IP 38. Video Frame Writer Intel FPGA IP 39. Video Streaming FIFO Intel® FPGA IP 40. Video Timing Generator Intel® FPGA IP 41. Warp Intel® FPGA IP 42. Design Security 43. Document Revision History for Video and Vision Processing Suite User Guide

30.3.2. Coefficient Quantization

The scaler IP implements the filters when you select Polyphase algorithm using fixed-point logic, so you must supply the filter coefficients in a fixed-point format. You define the format for the coefficients with parameters that select whether they are signed or unsigned, the number of integer bits, and the number of fraction bits.
  • Signed or unsigned: if you want to represent negative coefficients, turn on Use signed vertical coefficients. If all coefficients are positive values, reduce logic and turn off Use signed vertical coefficients.
  • Integer bits: the number of integer bits defines the maximum value that can be represented.
  • Fraction bits: the number of fraction bits defines the precision with which the IP can convert floating-point coefficients into the fixed-point format.

The overall bit width of each coefficient is the sum of the integer and fraction bits, plus one extra bit for signed coefficients. When using Lanczos coefficients, Intel recommends the following settings:

  • Turn on Use signed vertical coefficients as the Lanczos function for any number of lobes greater than 1 requires negative values, so the coefficients must be signed.
  • Use 1 integer bit as the maximum value required for any Lanczos coefficient is 1.0
  • Use between 6 and 8 fraction bits.

Typically, the filter coefficients produce noninteger floating-point values. To convert each floating-point coefficient into its closest quantized representation in the selected fixed-point format:

  • Multiply each coefficient by 2 frac , where frac is the number of fraction bits you select
  • Apply float to integer conversion to each coefficient

However, small errors in the coefficient values introduced by the quantization process can accumulate so that the coefficients in each phase no longer sum to their intended value. Generally, the coefficients in any phase should sum to exactly 1.0. Any value greater than 1.0 increases the overall brightness of the resulting image. Any value less than 1.0 reduces the brightness. The coefficients can sum to more or less than 1.0 if you want a brighter or darker image. You should still ensure your coefficients sum to your original, intended value post quantization. To restore the coefficients to values that sum to the intended value:

Figure 64. Restore Coefficient Values
float quantization_error = 0.0;
for (int j = 0; j < taps; j++) {
   quantization_error += original_float_coeff[j] - ((float)quant_coeff[j]);
   if (quantization_error < -0.5) {
      quant_coeff[j]--;
      quantization_error += 1.0;
   } else {
      if (quantization_error > -0.5) {
         quant_coeff[j]++;
         quantization_error -= 1.0; 
      }
   }
}