Compression Measures

AI Summary

An AI-generated summary of this video lecture.

This lecture introduces compression measures used to quantify the efficiency of data and image compression. The instructor begins by defining the Compression Ratio (CR) as the ratio of uncompressed file size to compressed file size, expressed as CR = N1/N2. Three compression cases are presented: no compression (CR = 1), significant compression (CR approaches infinity), and data explosion or reverse compression (CR approaches zero). A numerical example demonstrates how to calculate a 5:1 compression ratio. The lecture then transitions to Relative Redundancy (RD) and Saving Percentage (SP), defining RD as 1 - 1/CR or 1 - N2/N1, and SP as [1 - N2/N1] x 100. The instructor emphasizes that higher RD indicates greater compression, while RD = 0 means no compression and RD < 0 indicates data explosion. Bit Rate (BR) is introduced as the compressed data required per pixel. A worked example applies these formulas to a 256 x 256 pixel gray-scale image (8 bits/pixel) with an original size of 65,536 bytes compressed to 6,554 bytes. The example calculates CR approximately 10:1, RD approximately 0.90, SP approximately 90%, and BR approximately 0.80 bits/pixel.

Chapters

1. 0:00 – 2:00 00:00-02:00

   The lecture opens with a slide titled 'Compression Measures' that defines the Compression Ratio (CR) as CR = Uncompressed file size / Compressed file size = N1/N2. The instructor annotates the formula with 'N1 (Original)' and 'N2 (Compressed)'. Three color-coded boxes illustrate compression cases: NO COMPRESSION with N1 = N2 and CR = 1, SIGNIFICANT COMPRESSION with N2 << N1 and CR approaching infinity, and DATA EXPLOSION / REVERSE COMPRESSION with N1 << N2 and CR approaching zero. A handwritten numerical example shows 10/2 = 5:1 to demonstrate ratio calculation.

2. 2:00 – 5:00 02:00-05:00

   The instructor continues explaining the compression ratio formula and its three cases, drawing red circles around key values such as '2' in the example. The lecture then introduces Relative Redundancy (RD) with the formula RD = 1 - 1/CR or RD = 1 - N2/N1, noting that higher RD indicates greater compression. Saving Percentage (SP) is defined as SP(%) = [1 - N2/N1] x 100, representing the percentage of storage saved. Bit Rate (BR) is introduced as compressed data required per pixel. The instructor highlights that RD = 0 means no compression and RD < 0 indicates data explosion, using red boxes to emphasize these conditions.

3. 5:00 – 10:00 05:00-10:00

   A worked example slide presents an image-compression problem: a 256 x 256 pixel gray-scale (8 bits/pixel) file of 65,536 bytes compressed to 6,554 bytes. A GIVEN box lists N1 = 65,536 bytes, N2 = 6,554 bytes, gray-scale = 8 bits/pixel, and Number of Pixels = 256 x 256. Four numbered panels compute the measures: CR approximately 10:1, RD approximately 0.90, SP approximately 90%, and BR approximately 0.80 bits/pixel. A FINAL ANSWERS bar repeats CR = 10:1, RD = 0.90, and SP = 90%. Red handwritten annotations include '1P = 8' and circled values for each computed measure.

4. 10:00 – 10:02 10:00-10:02

   The final frames show the completed worked example slide with all four compression measures calculated and annotated. The instructor's red handwriting marks key values including CR = 10:1, RD = 0.90, SP = 90%, and BR = 0.80 bits/pixel. The GIVEN box and FINAL ANSWERS bar remain visible, providing a complete reference for the example problem.

The lecture systematically builds understanding of compression measures from basic definitions to applied calculations. The central concept is the Compression Ratio (CR = N1/N2), which establishes a baseline for measuring compression effectiveness. The three cases (no compression, significant compression, data explosion) provide a conceptual framework for interpreting CR values. Relative Redundancy (RD = 1 - 1/CR) and Saving Percentage (SP = [1 - N2/N1] x 100) extend this framework by expressing compression efficiency in normalized and percentage terms, respectively. The worked example demonstrates how to apply all four measures (CR, RD, SP, BR) to a realistic image-compression scenario, reinforcing the relationships between these formulas. Key exam-relevant points include: CR > 1 indicates compression, CR = 1 means no change, CR < 1 means data explosion; RD ranges from negative (data explosion) to approaching 1 (maximum compression); SP and RD are numerically equivalent when expressed as percentages; BR = N2 x 8 / number of pixels for byte-based data.