CSCI130

1
Lecture 3 (1.1-2.3)

• CSCI130
• Instructor Dr. Imad Rahal

2
Non-numeric Data Representation

• Text data
• The most common type of data people use on
  computers
• How can we transform it to binary --- not as
  intuitive as numbers!
• Numbers were (relatively) easy to map to binary
• Decimal to binary change
• Words can be divided into characters
• Each character can then be encoded by some binary
  code
• Every language has its set of letters ? we will
  limit ourselves to the Latin alphabet
• What about letters and other symbols?

3
Non-numeric Data Representation

• Many transformations exist and all are arbitrary
• Two popular
• EBCDIC (Extended Binary Coded Decimal Interchange
  Code) by IBM
• ASCII (American Standard Code for Information
  Interchange) by American National Standards
  Institute (ANSI)
• Most widely used
• Every letter/symbol is represented by 7 bits
• How many letters/symbols do we have in total?
• A-Z (26) , a-z (26) , 0-9 (10), symbols (33),
  control characters (33)
• If using 1 byte/character ? we have one extra bit
• Extended ASCII-8 (more mathematical and graphical
  symbols or phonetic marks)
• Why are passwords case sensitive?
• How to sort alphabetically

4
(No Transcript)
5
Non-numeric Data Representation

• Given a memory register with the value 00110100
• 2s complement ? 52
• Floating point ? 0.625
• ASCII ? character 4 (check ASCII table in book)
• There are some code blocks preceding such values
  informing the computer of the type
• Sometimes called meta-data
• E.g. Font style

6
Non-numeric Data Representation

• Picture/Image Data
• 512256 image ? grid has 512 columns and 256 rows
• Divide the screen into a grid of cells each
  referred to as a pixel
• Pixel values and sizes depend on the type of the
  image
• Black white images 1 bit for every pixel such
  that 1 is black and 0 is white
• Grayscale images 1 byte/pixel where 255 is black
  and 0 is white and anything in between is gray
  (higher/lower values are closer to black/white)
• Color images Three values per pixel
• Red/Green/Blue
• 1 byte per color ? 3 bytes per pixel
• For an 512x256 image (512x256x3 384K bytes)
• For any image, we only store the pixel values

7
(No Transcript)
8
An Image File

• //B/W IMAGE
• P1
• 15 11
• 0 1 1 1 1 0 1 1 1 1 0 1 1 1 1
• 0 1 0 0 0 0 1 0 0 0 0 1 0 0 1
• 0 1 0 0 0 0 1 1 1 1 0 1 1 1 1
• 0 1 0 0 0 0 0 0 0 1 0 1 0 0 1
• 0 1 1 1 1 0 1 1 1 1 0 1 1 1 1
• 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
• 0 1 1 1 1 0 0 1 1 1 0 1 0 0 1
• 0 1 0 0 0 0 0 0 0 1 0 1 0 0 1
• 0 1 1 1 1 0 0 0 0 1 0 1 0 0 1
• 0 0 0 0 1 0 0 1 0 1 0 1 0 0 1
• 0 1 1 1 1 0 0 1 1 1 0 1 1 1 1
• //COLORED IMAGE
• P3
• 2 7

9
Non-numeric Data Representation

• Image movies are built from a number of images
  (or frames) that are displayed in a certain
  sequence at high speeds
• 30 images per second
• Assume every image has a size of 500KB
• 500K 30 15 MB
• 2-hr movie needs (assume same sample image used)
• 15MB 60 120 108 GB (billion) bytes!

10
Non-numeric Data Representation

• People use compression to reduce large movie
  sizes
• (Temporal compression) Usually the change between
  two consecutive images is small ? store only
  difference between frames

11
Non-numeric Data Representation
Heard sound depends on the amplitude and
frequency

• Sound/Audio Data
• Produced when objects vibrate in matter (e.g.
  air)
• Sends a wave of pressure fluctuations
• Sounds differ because of variations in the sound
  wave frequency (pitch or speed)
• Frequency cycles/second
• Higher wave frequency ? pressure fluctuation
  switches back and forth more quickly during a
  period of time
• We hear this as a higher pitch
• Level of air pressure in each fluctuation, the
  wave's amplitude or height, determines how loud
  the sound is

Volts
12
Non-numeric Data Representation

• Not all frequencies are audible
• Your ears are particularly sensitive to sounds in
  the middle range, from about 500 Hz to 2 kHz
• The hi-fi range is defined as from 20 Hz to
  20 kHz
• As you get older, you will find it more and more
  difficult to hear higher frequencies
• By the time you are able to afford a decent hi-fi
  system, you will probably be unable to fully
  appreciate its performance ?

13
Non-numeric Data Representation

• Numbers used to represent the amplitude of sound
  wave
• Analog is continuous and we need digital
• Digitize the sound signal
• Measure the voltage of the signal at certain
  intervals (e.g. 40,000 per sec for CDs)
• process of sampling
• Reconstruct wave
• Sound will slightly vary
• A sound file is nothing but a sequence of numbers
  measured at equal intervals

14
Non-numeric Data Representation

• Higher frequency/pitch
• more cycles during same time interval
• less time between any two cycles
• less measurements between any two peaks (given
  that measurements are taken at equal intervals of
  time)
• Compression can also be used for audio files
• MP3 reduces size to 1/10th
• ? faster transfer over the Internet

15
Non-numeric Data Representation

• Digital image and audio have a lot of advantages
  over non-digital ones
• Can easily be modified by changing the bit
  pattern
• Image enhancement, noise/distortion removal, etc
  
• Superimpose one sound on another or image on
  another results in newer ones
• Not admissible as evidence in courts of law

16
Non-numeric Data Representation

• adapted from a course offered at BU

17
Non-numeric Data Representation