cs525 logo
rainbow animatio

Solution to Homework #3: Video Processing, DCT, and JPEG Encoding


  • Learn basic video processing.
  • Learn Discrete Consine Transformation.
  • Learn JPEG entropy coding.
  • Assignment Date: 3/1/2010

    Due Date: 3/8/2010

    Related References:



    1. In year 2009, ATSC decides to start developing the next generation of HDTV standards with four times the resolution of today's HDTV system. Assume one simple solution is to increase the scanning lines from 1125 to 2250 and increase the BW 4 times, other paramters (CH, AR, FR, CV, Kf) not changed. What will be the RH and Rv numbers? What will be number of equivalent pixel in one frame? See pages 12 and 13 of video.pdf for the related formula.


      NTSC PAL HDTV hw3p1HTV
      BW 4.20 5.00 20.00 80.00
      NL 525.00 625.00 1125.00 2250.00
      CH 0.85 0.80 0.83 0.83
      AR 1.33 1.33 1.78 1.78
      FR 29.97 25.00 30.00 30.00
      RH 340.34 384.00 553.33 1106.67
      Cv 0.92 0.92 0.96 0.96
      NL 525 625 1125 2250
      Kf 0.7 0.7 0.7 0.7
      Rv 338.1 402.5 756 1512
      Np 153425.4 206080 743680 2974720

      See the excel spreadsheet at http://cs.uccs.edu/~cs525/hw/solution/hw3/hw3p1q1.xls
      You can use the spreadsheet in midterm to anwser the related questions.
    2. What is the main difference between a composite TV system and a component TV system? (See Page 25 of video.pdf)
      Ans: Component TV signal transmits the RGB signals separately. Therefore they are preserved better than the composite TV signal where the YIQ signal components are mixed and transmitted over a single channel.
    3. The video standards converts RGB signal to YIQ signal format. how can we take advantage of the YIQ format, in terms of video compression utilizing the human perception system? (See Page 26 of video.pdf).
      Ans: Since human eye is more sensitive to luminance than I signal (orange-cyan color), and least sensitive to Q signal (green-magenta color), we can use more bits to encode the Y (luminance) signal, fewer bits to I signal, and fewest bits to Q signal. Or we can dither the image plane of I signal and Q signal further so that fewer pixels (or equivalent of fewer bits are used to present them.)