CHAPTER ONE NOTES
- mm = multimedia
- dmm = digital multimedia
- digital multimedia can be interactive
- www became publicly avail outside CERN start if 1992,in form of line based browser, access to handful of servers. line based browser displays text rather than graphics(icons). ex of lined based browser is Lynx
- 1st films exhibited to public by Lumiere Bros. in Paris 1895
- mixture of media under software control...where display & presentation of the media elements is the sole purpose
- multimedia production, where media is more intimalely bound up with computation
- modalities, various perceptions of reading text, listening to sound, looking at pictures in multimedia
- multiple media requires us to switch between modalities. ex. reading text then looking at a picture
- digital mulimedia, combo of 2 or more media, represented in a digital form, well intergrated to be presentd via a single interface, or manipulated by a single computer program.
- 1995 industry consortium announced for successor to the CD-ROM using same size and shape platter...DVD (digital versatile disk) capacity up to 17 Gbytes on a double sided disk.
- generally when mulitimedia is del online the del need not be passive
- 2 models currently in use for combining elements of different media types...PAGE BASED & TIME BASED
- time based elements are film, audio video. emdedded in a page as if they were images occupying a fixed area
- page base elements are books and world wide web
- linked page based multimedia productions are known as hypermedia the best know ex. of this is the world wide web
- time based multimedia makes time the central organizing principle
- parallelism, several video clips shown at the same time perhaps overlaid against a static image or a sound track may play during an animation
- flash, most widely used time based mm (multimedia) technology on the web. jumps between frames controlled in interactive branching and loops
- powerpoint slideshows provide the most basic sor of mm presentation
- non linear books...dictionary, phonebook, cookbook, thesaurus
- interactivity..portrayed as the feature that differs dmm from other forms of comdined media.
- interactivity empowers the end user of your project by letting them control the content and flow of info to an extent.end users can control only choices that are coded into the program that are allowed
- ultimate control over "content and flow of info " remains with the producer
- user interfaces, extreme one..menus, dialogue boxes, outline buttons and conventions used by most operating systems, mainsteam apps and web forms. Extreme two..interaction with some sort of game is free form w/ any part of the screen liable to cause a response at some time.
Social and Ethical Considerations
- new technology creates new opportunities for new unethical behavior
Access to MM: Consumption
- highest level of computer skills are usually acquired through work
- physical disabilities or learning difficulties may interfere with a persons ability to use a computer
- examples....arthritis, motor neurone disease, blindness, dyslexia, color blindness
- available devices to help these issues..voice synthesizers, braille keyboards and monitor color combinations
Control of MM and Content
- 1999 Apple launched quicktime tv, providing video over the internet
- platform for internet content selection (pics), the difference between pics & conventional mechanisms, such as banning books or siezure of video tapes is it restricts reception , not distribution
- internet content rating association(icra) independent org which admins the labelling of web sites, does not do it itself but through a questionaire provided to the web master.
- icra perceives its mission as protection of children , at the same time it wishes to prserve free speech.
Slide Notes Chapter 1
- affordances is a term from ecological psychology
- different things afford different interaction
- consider door knobs, levers, push-plates and pulls
- Don Norman and Psychology of Everyday Things
- scroll is read in linear fashion. it is not page based media
- internet access is not universal
- william randolf hearst..newspaper proprietor who owned international film service and syndication rights to popular comic strips.
CHAPTER TWO NOTES
- computers are built to only have 2 states, on and off
- computers store and operate on BITS
- bits only have 2 values "0 and 1" they can be read as #s to BASE 2
- bytes are 8 bits and arranged in a linear sequence so that each byte can be identified by its position in the sequence. THIS IS KNOWN AS THE ADDRESS
- bit patterns can represent INSTRUCTIONS that cause the processor to carry out operations on values stored in memory.
- instructions = bit patterns therefore sequences of instructions are PROGRAMS that can be stored in memory and executed.
- STORED PROGRAM MACHINE- the defining characteristic of a computer.
- Two coding schemes used Hexadecimal & ASCII
- Hexadecimal: 16 fingers, 16 digits
- Humans organize 0s and 1s into groups of 4
- These groups of 4 are can be represented by a single hexadecimal digit
- 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F
- ASCII made of two hexadecimal codes
- One ASCII character - two hex codes
- ASCII code for R (from text pg 317)
- hexadecimal: 52
- binary: 0101 0010
- How to count using a different number of fingers
- 10 fingers: Counting in decimal
- 0, 1, 2, 3, 4, 5, 6, 7, 8, 9,
- start over but put a 1 in the higher position
- 2 fingers: Counting in binary
- 0, 1
- start over but put a 1 in the higher position
- EXAMPLE: 100,101,110,111,1000
- 16 fingers: Counting in hexadecimal
- 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F
- start over but put a 1 in the 1 higher position �
- EXAMPLE: 01,02,03,04,05,06,07,08,09,0A,0B,0C,0D,0E,0F,10,11,12,13,14,15,16,17,18,19,1A,1B,1C,1D,1E,1F,20,21,22,....
- it is frequently necessary to convert from analogue representation to a digital one, ex. by scanning a picture
- digitization- converting a signal from analogue to a digital one
- digitization consist of 2 steps..SAMPLING (measuring the signals value at discrete intervals) and QUANTIZATION (when we restrict the value to a fixed set of levels)
- analogue to digital converters(ADC's) hardware device
- sampling rate-# of samples in a fixed amount of time or space
- quantization levels-the level that a signal is quantized
- advantage that digital rep have over analogue ones stems from the fact that only certain signal values "those over quantization levels" are valid
- if a signal is transmitted over a physical medium, such as a magnetic tape, inevitably random NOISE is intro
- noise will cause a signal value to change.
- if signal is digital any minor fluctuations caused by noise will make a legal value into an illegal one that lies between quantization levels. restore the original signal by quantizing agian
- sample and hold-the value of a sample is used for the entire extent between it and the following sample
- undersamplingresults in inadequate quality from the original "manifestted as distortions and artifacts which are always undesirable
- harmonics-the # of overtones present in a note played by an instrument that gives each instrument its own TIMBRE
- timbre -
- the collection of frequencies and amplitudes is the signals representation in the FREQUENCY DOMAIN
- frequency domain can be computed using a mathematicl operation known as the FOURIER TRANSFORM
- DC component is a spike at a frequency of zero
- higher frequency components are associated w/ abrupt transitions.
- as higher frequencies are added, the leading & falling edges of the waveform become more nearly vertical
- filters that remove certain frequencies - sharpening or smoothing an image
- inverse fourier transform - taking a signal from the frequency domain to the time domain
- Sampling Theorom - if the highest frequency component of a signal is at Fn, the signal can be properly reconstructed if it has been sampled at a frequency greater than 2Fn. This limiting valua is known as the NYQUIST RATE
- posterization when a region of an image w/ a continuous gradation of tone is replaced w/ several regions of fewer tones resulting in an abrupt change from one tone to another
- The most common reason for limiting the number of quantization levels (in any type of signal) is to reduce the amount of memory occupied by the digitized data by restricting the number of bits used to hold each sample value.
- quantization noise - when sound is quantized to too few amplitude levels, the result is perceived as a form of distortion. *Quantization noise is clearly discernible when sound is sampled to 8 bits (256 levels), but not (except in the opinion of some audiophiles) at the 16 bits (65536 levels) used for audio CDs.
- Hardware & Software Requirements
- When looking at the hardware requirements of multimedia, there are two distinct cases to consider: requirements for multimedia consumption and the requirements for its production.
- (MPC) multimedia pc
- (PDA) personal digital assistant
- pda's are based on cellular telephone networks which provide LESS bandwiththan a fixed network
- raw processing power, high speed data buses, large main memeory, & pwerful graphics boards are a necessity for producing mm content; fast high secondary storage is equally important
- RAID-Redundant Array of Inexpensive Disk. put together out of cheaper and slower disk. improvement is achieved by reading and writing to them in parallel
- level 0 RAID, the lowest level, specifies a "data striping technique." a data block being written to the array is split into segments which are written to seperate disk.
- RAID 0 offers no protection against disk failure, if one disk fails the array will fail
- level 1 RAID, the least sophisticated, mirrors disk which decrease production
- combine RAID levels 0 an 1 will allow you to gain performance of data striping and the protection of disk mirroring
- the key to integration is a framework that can accomodate a multiplicity of media and present them to the user. there are 3 approaches....WWW, API, and FLASH
- WWW uses 2 computer languages...HTML(hypertext markup language) or XML(extended markup language)
- (API) application programming interface, example QUICKTIME
- flash, provides mm production in "stand alone" form, no need for additional software to be used
- LAN'S are connected together by routers, bridges, and switches to form internets.
- the internet is a global network of networkscommunicating by a standard set of protocols loosely referred to as TCP/IP
- TCP/IP- transmission control protocol / internet protocol. internet rules of communication, who speaks first
- 56kbits per second- the max bandwith avail through a modem
- broadband - bandwith connectivity 500 kbps or greater. examples.. ADSL, cable modem, satellite
- asymmetric digital subscriber line
- over copper phone wires
- limited to short distance from phone switch
- 6.1 mbps (million bps) downstream
- 640 kbps upstream
Clients and Servers
- mm distribution over lans or internet based on client/server model of distributed computation
- server always"listens" and responds
- client "request"or consumes
- WWW clients are browsers such as mozilla, firefox, internet explorer
- protocol - rules governing actions taken by a server/client when it receives a request or response
- WWW - most popular form of mm online delivery
- HTTP - simple protocol, for fast transmission of hypertext
- info in the WWW is described as HYPERMEDIA
- URL (uniform resource locator) "the familiar web address
- FTP (file transfer protocol)- one of the TCP/IP family of protocols used for transferring files
- IP address the name maps to a number
- MIME (multipurpose internet mail extension)- extension to the internet mail protocols that support the inclusion of data other than plain ASCII text in mail messages
- MIME allows the transmission of more than just ASCII text (like you’d expect in an email)
- MIME types are specified in the header
- Huge variety of MIME types are allowed
- audio, images, video
- compressed files
- IANA (internet assigned numbers authority)
- ISO (international organization for standardization)
- "Standards are documented agreements containing technical **specifications or other precise criteria to be used consistently as rules, guidelines, or definitions of characteristics, to ensure that materials, products, processes and services are fit for their purpose."
- Multimedia standards are used for: interfaces, file formats, markup languages, network protocols, etc.
- Three international standard organizations:
- The ISO
- The International Electrotechnical Commission (IEC)
- The ITU (International Telecommunication Union)
- ISO does every technical field besides electronic engineering (goes to the IEC)
- ITU is an agency of the UN.
- ad hocapproach accommodates rapid change
Slide Notes Chapter Two
- .doc, .jpg,.xls are ex of file extensions
- file extensions may not be visible, buried in the file itself and computer reads
- discrete vs continuous
- Things in the real world can be discrete
- They either ARE or ARE NOT there
- These things can be counted
- The number of cars in the parking lot
- The number of beans cin a jar
- discrete vs continuous
- Things in the real world can be continuous
- Continuous can’t be counted, it must be measured
- Atmospheric pressure
- Height of an ocean wave
- Frequency of a sound wave
- computers can only count
- Discrete data is easy for a computer
- count it and store it as a number
- Continuous data... not so much
- measure the frequency & amplitude
- encode as discrete
- measure the amount of light and its color
- encode as discrete
CHAPTER THREE NOTES
Vector Graphics & Bitmapped Graphics
- pixels- monitors display pictures as rectangular array pixels. Small, usually square dots of color which merge optically when viewed. pg66
- low level operations required to set pixel values are performed by a graphic library, which communicates with the display hardware and provides a higher level interface to the application program. pg66
- a graphics application program must somehow keep an internal model of the image to be displayed. pg66
- rendering- process of generating a pattern of pixels from a model.pg66
- Internally an application keeps a numeric model
- Externally an application must project this internal model onto a bitmapped **display
- An application must also project this internal model onto a storage model to be able to store and recall it later
- Moving from logical and physical pixels
- called rendering
- may be different size and different resolution
- will probably require clipping and scaling to move from logical to physical pixels
- two different approaches to graphic modeling: bitmapped graphics and vector graphics. pg67
- bitmapped graphics- the image is modeled by an array of pixel values. pg67
- grandfathered name... more like pixel mapped graphics
- logical pixels (store values) correspond one to one to physical pixels. pg68
- images are modeled internally as an array of pixel values... the logical pixels
- physical pixels- physical dots on the screen. pg68
- scaling and clipping are the only computations that need to be performed to display a bitmapped image. pg68
- Scaling and Resize
- Vector? Simple... change formula
- Changes can be made BEFORE pixel values are calculated
- Bitmapped? Complicated...
- frequently results in artifacts (artifacts-unwanted side effect to data/image)
- vector graphics- the image is stored as a mathematical description of a collection of individual lines, curves and shapes making up the image. pg68
- a line is defined by slope, length & position
- a circle is radius & position
- a square is length of side & position
- Rendering is very different w/ vector graphic
- must translate equations to physical pixels
- not simply clip or scale
- must compute the array of physical pixels from the equations
- a bitmapped image must record the value of every pixel, a vector description can be much more compact. pg68
- Bitmapped image size is set by
- size, resolution and color resolution
- not affected by contents
- Vector graphics size is set by
- the contents of the image
- the more complex, the larger the file gets
- resolution does not affect the size of the file
- painting programs operate on bitmaps. pg 70
- drawing programs work with vector representations; adobe illustrator or freehand, intendend for creating or editing vector graphics, and only provide limited facilities for dealing with bitmaps. pg70
- photoshop and painter provide extensive support for bitmap manipulation. Either offer little support for vector graphics.pg70
- Major difference between vector and bitmap is the way the behave when scaled or resized. pg70
- Bitmap image, if displayed larger than its natural size, each logical pixel must be mapped to more than one physical pixel on the final output device. Acheived by multiplying up the logical pixels or by interpolating new pixels. The effect will usually be loss of quality. pg71
- vector image, scaling can be performed easily as a simple mathematical operation before the pixel values are calculated. Curves will remain smooth no matter how much a vector image is blown up. pg71
- bitmap images become jagged or blurred if scaled up. pg71
- drawing program (vector) lets you build up a vector image out of individual curves lines and shapes. Usually has much of the character of a pen and ink illustration. pg72
- painting program (bitmap) allows you to make a wide range of different marks and apply color to arbitrary areas of the image. pg72
- painting programs (bitmap) offer more expressive possiblities but at the expense of high memory requirements and scalablity problems. pg72
- scanned images, screen shots, photos from digital camera and captured video frames are uaually all bitmaps. pg72
- charts, diagrams and other data visualizations generated by a program from data usually use vector graphics. pg72
Combining Vectors and Bitmaps
- vector graphic into a bitmap image easily done pg73
- rasterizing- the process of intrepreting the vector description. The rasterized image looses all of its vector properties. pg73
- Rasterized image ceases to be resolution independent. A resolution must be chosen when the rastorization takes place. pg73
- Converting pixels (bitmap) to vectors is much more difficult. It requires software to identify the boundaries of shapes within the image. Then to approximate those boundaries using available curves and lines and to color them appropriately. pg73
- Vectorization can be used in a controlled fashion to generate a starting point for a new vector image. Most drawing programs allow you to allow bitmaps without vectorizing them. A bitmap imported in this way is treated as an invisible object. pg74
- The arrangement of artwork in layers is an organizational device that is common to both vector and bitmap images. pg75
- Layers can be duplicated and the duplicates altered separately. pg77
- A layer is often compared to a digital version of a sheet of clear acetate material, like an overhead projector transparency. pg75
- used in photoshop heavily
- Bitmapped and vector graphics use this as an organizational device
- In bitmapped graphics
- layers are used like digital tracing paper to isolate objects in the image
- colors can be separated and manipulated individually
- Lossless compression algorithms have the property that is always possible to reconstruct the original data exactly from its compressed version. pg79
- Lossy algorithms discard some data, in the case of images data representating visually insignificant details in order to achieve greater compression. pg79
- GIF, developed by CompuServe, for exchanging bitmapped images between different platforms. Uses a lossless compression technique and restricted to 256 colors. Most suitable for simple images such as cartoon style drawings and synthetic images produced on a computer. pg79
- JPEG (Joint Photographic Experts Group)- JPEG is a compression technique, images that have been compressed using it maybe stored in any of several file formats.pg80
- JPEG data can be imbedded in other files.pg80
- TIFF (Tag Image File Format)- is an elaborate extensible file format that can store full color bitmaps using several compression schemes. pg80
- BMP- is more properly called the Microsoft Windows Bitmap Format. It only supports a simple form of lossless compression and are usually store uncompressed. pg80-81
- SVG (Scalable Vector Graphics Format) pg82
- EPS, SWF, and SVG are not just vector formats. Their main feature is vector capablities however it is possible to incorporate bitmaps into these files as self contained objects.
- Graphics metafile- a format that accomadates both vector, text and bitmap graphics. pg82
CHAPTER FOUR NOTES: VECTOR GRAPHICS
- The contents of a vector graphic affects its size more than it affects the size of a bitmapped graphic.
- Provide and elegant way of constructing digital images whose representation is compact, scaleable, resolution-independent, and easy to edit. pg 86
- Vector Graphics are attractive for networked multimedia because of their compactness.pg 86
- mandatory for 3-D graphics
- AKA vectors - vectors have magnitude (length) and direction, (from one point to the other)
- In vector graphics, images are built up using shapes that can easily be described mathematically. pg 86
- When studying vector graphics, it is important to be familiar with the coordinate geometry, which is the field of mathematics underlying the representation of shapes in vector graphics.pg 86
Coordinates and Vectors
- coordinates-any pixel uniquely identified by the pair (x,y) where 'x' is the column # and 'y' is the row #. example (7,3) 7 is 'x' (the column) and 3 is 'y' (the row) pg 87
- vector graphics coordinates (real values), A point is defined by its x and y coordinate
- Can be fractional, Can be negative
- coordinates of pixels in an image must be integer values (whole numbers) between zero and the horizontal (for x coords) and vertical (for y coords) deminsions in the image pg 87
- rulers-axes along the edges of your drawings, displayed in vector drawing programs. pg 87
- coordinate transformation-cooridnates in one system (the user space) are transformed into a different one( the device space). one example would be when an image is rendered in a window on a display.pg 87 & 88
- pairs of coordinates are used to define points and also used to define displacements pg 88
- displacements - distance between points or movement from one point to another and can be defined by a pair of points
- a coordinate system allows us to identify points in space pg 88
- coordinate geometry comes from using letters to represent 'unknown' values and using equations in those values to specify relationships between coordinates that characterizegeometrical shapes. pg 88
- pixel coordinates are always whole numbers, you cannot set a value for just part of a pixel.
- Bounding Box - images can be contained inside a “bounding box” which is the smallest box that contains all the points found in an object, a bounding box is placed by its (0,0) position
- Absolute coordinates - are coordinates in relation to the entire screen
- Window coordinates - distance measured inside the window from upper left hand corner. window coordinates are within the window you are looking at
- When we view a digital photograph, the reconstruction (interpolation) is performed by a display or printer device, and by our eyes and our brain. If the reconstructed image differs from the original image, we are seeing an (ALIAS) i.e. Moire’ patterns
- Anti-aliasing- the replacement of pixel values at a contrasting edge in an image (especially a diagional edge) with values between the original extremes. This softens the coarse step like appearance (pixellation) often seen in low resolution images. pg 90
- Scaling a bitmapped graphic sometimes causes the edges of lines to look jagged. A technique that makes those edges appear to be less noticable is called
- sampling and reconstruction- the process of rendering a vector object to produce an image made up of pixels pg 89
- jaggies (artifacts) are a form of aliasing caused by undersampling pg 90
- if an image contains a sharp hard edge boundary, its brightness and color will change directly from one value to another crossing the boundary without any intermediate gradation.pg 90
- no sampling rate will be adequate to ensure perfect reconstruction, jaggies are always possible no matter how high the resolution that is used for rendering a vector shape. pg. 90
- Bezier Curve - a curve defined by 2 end points with a pair (2 more points called direction points)of direction lines at each end to indicate the directions and rates at which the curve leaves the two end points. a class of smooth curves.
- Shapes can be filled with color, patterns or gradients. pg 91
- polyline - a sequence of connected lines, sometimes considered a single object.pg 91
- closed polylines- first and last points coincide, form regular or irregular polygons, can be used to draw rectangles,(they form a shape)
- Bezier Curve break down... a total of 4 points,(P1,P2,P3,P4) 2 end points and 2 points called direction points which usually do not lie on the curve itself. pg 92
- control points - the end points and direction points on a bezier curve pg 92
- the length of the lines from each end point to its direction point determine how wide a sweep the curve makes. pg 93
- direction line - a line, usually when using a pen tool, showing you how far you have pulled the a line from the end point.
- path-collection of lines and curves
- closed path -if the path joins up with itself, there are no end points pg 96
- open paths have end points pg 96
- each individual line or curve is called a segment of the path pg 96
- anchor point - the points where segments join (the original endpoints of the segment) pg 96
- any collection of curves and lines may be considered as a path, they do not all need to be connected pg 96
- the usual way to construct a path is by costructing the individual segments with the pen tool which provides great control and permits you to construct accurate curves. pg 96
- a path is an abstract mathematical entity pg 96
- stroke and fill - making a path visible by either applying a stroke (as with ink or some other medium) to the path or you treat it as the outline shape and fill it (as if with paint or ink)or you can do both. pg 96
- three line cap options provided by postscript....butt cap, round cap, projecting cap pg 97
- butt cap - squared off end, may produce an undesirable and ugly effect pg 97
- round cap - preferred, line filled with a semicircle built across the end pg 97
- projecting cap - a stroke continued beyond the endpoint of the path by half the width, so that the weight of the stroke relative to the path is the same in all directions. pg 97
- three styles of line joining provided be Illustrator...mitre, round, bevel pg 97
- mitre - (picture frame) the outside edges of the lines are extended to meet at a point pg 97
- round - a circular arc is used to produce a rounded corner
- bevel - the segments are finished off square where they join, and the resulting notch is filled in with a triangle to produce a flat ended joint. pg 97
- you can only fill a closed path, however, most drawing programs also allow you to fill an open path...the filling operation implicitly closes the path with a straight line between end pointspg 98
- a single color is the simplest fill pg 98
- Gradient fills - are widely used in artwork created in vector drawing programs, and contribute to the characteristic air-brush look of graphic design pg 98
- linear gradient - colors at each end of a region are specified, and a smooth blend of intermediate colors is generated between pg 98
- radial gradient - color varying outwards from a center point to the outside of the fill pg 98
- tiles - a small piece of artwork, made using the facilities provided by your drawing program ( much like piecing together bathroom tile). pg 99
- pattern fills (such as tiles) are often used as backgrounds pg 99
- non zero winding rule - used to determine whether a point is inside a path. Greater explanation on how it works pg 100
Transformations and Filters
- Translation- moving the shape from one place to another pg 101
- Scaling- changing the shape's size pg 101
- Rotation- rotating the shape around a point (must identify the rotation point (axis) and what you rotate the object around) ( pg 101
- Reflection- changing the shape's position so that it mirrors the original one (across a line) pg 101
- Shearing- changing the angles of the shape pg 101
- filter -a program in Illustrator and similar programs that add effects to to an object. Examples are.... roughening (produces a rough edge to an object), scribbling (moves anchor points in random fashion),rounding corners (converts corner points into smooth curves) pg 102
- transformations are simply achieved by altering the coordinates of the defining points of objects. pg 102
- 3-D shapes (objects) are defined by their surfaces
- Made even more complicated by the fact that a 3-D object inside the computer must be translated into 2-D to be rendered on a computer screen...this results in the need to specify the viewpoint.
- pictures on a screen are always two-dimensional pg 103
- 3-D graphics - vector graphics based on three deminsions pg 103
- 3 dimensions,(x, y and z)..x & y form the ground plane..Z is the height
- Geometrical shapes of 2-D are replaced by 3-D objects: instead of a circle, 3-D is a sphere, instead of a square, 3-D is a cube pg 103
- complex objects have a hierarchical structure...sub objects, sub sub objects and so on. pg 105
- In 2D the axis of rotation is always perpendicular to the X Y axis
- 3D rotation is more complicated than 2D since you must specify an axis of rotation.
- 3-D: additional complexity
- surface texture
- rendering is extremely computationally expensive (demanding)
- constructive solid geometry - the simplest approach to 3-D objects, uses a few primitive geometric solids such as the cube, cylinder, sphere, and pyramid to construct more complex objects. pg 106
- 3-D objects build by squishing and stretching those objects
- 3-D operations describing the combining of two complex (3-D) objects... union, intersection, and difference pg 106
- union - is a new object made out of the space occupied by the two together pg 106
- intersection - the intersection of two objects is the space that the two have in common pg 106
- difference - the difference of two objects is the space occupied by the the first but not the second pg 106
- Fractals are a mathematical idea that incorporates an infinite area with a finite perimeter. Example -Draw a triangle. Now draw one on top of it upside down. Keep doing this to infinity rotating a fraction of a degree over and over. pg 108
- free form modeling - uses a representation of an objects boundary surface as the basis of its model. pg 106
- free form modeling - Build surfaces from flat polygons or curved patches, results in an object drawn as a “mesh”
- Extrusion - Draw a 2 dimensional shape through space along a line, the line can be straight or curved
- procedural modeling-Objects are defined by algorithms or procedures, best known is based on Fractals
- Fractals- used to model natural objects,exhibit same structure at all levels, AKA "self similar"
- meatballs-soft objects that you can stick together
- 3-D models only exist in computers
- Radiosity is an attempt to model the complex reflections that occur between surfaces that are close together pg 114
- specular reflection- when the light that bounces off the surface of shiny objects.
- Ray tracing works by tracking the path of a ray of light back from each pixel in the rendered image to a light source. pg 114
- The simplest way of shading an object whose surface is made out of polygon is to calculate value for each polygon. pg 113
- rectangles and squares
- rectangles can be described by two corners
- squares are special cases of the rectangle
- ellipses and circles
- ellipses can be described by two points
- circles (1 point)and the radius; are special cases of the ellipses
- Structural Hierarchy
- Things in the real world are compositions of smaller things
- Things in the 3-D graphics world are also compositions of smaller things
- Hierarchical structure is an excellent way of coping with complexity
- see second half chapter 4 power point slide 12, 13, 14 for RENDERING 3-D.
CHAPTER FIVE NOTES: BITMAPPED IMAGES
- RESOLUTION is a measure of how a device approximates continuous images using finite pixels. pg 118
- resolution is closely related to sampling and some of the ideas about sampling pg 118
- 2 common ways of specifying resolution
- dots per inch....printers and scanners
- pixel dimensions - resolution is normally specified by giving the size of a frame, measured in pixels...example video
- image quality is dependent on pixel rate.
- VIDEO DISPLAYS image resolution in the form of pixel dimensions because the same pixel grid is used to display the picture on any monitor(using the same standard) regardless of its size. HOWEVER, monitor resolution is sometimes quoted in dots per inch because of the tendency in computer systems to keep this value fixed and to increase the pixel dimensions of the displayed image when a larger display is used. pg 119
- color printers complicated - arrange dots in groups using patterns of different colors within each group to produce the required color, SO the size of the color pixel is greater than the size of an individual dot of ink. pg 119
- LINES PER INCH - resolution of a printer taking account of the way of mixing colors pg 119
- screen ruling - the numbers of lines per inch will be as few as one fifth of the number of dots per inch pg 119
- bitmap images are an array of pixel values, they have pixel dimensions and unlike an input and output they have no physical dimensions. pg 120
- physical dimension = pixel dimension / device resolution ...the device resolution is measured in pixel per unit length. pg 120
- in order to maintain an images natural size most image formats record a resolution w/ the image data, this resolution is usaually quoted in units of pixels per inch(ppi) to distinguish it from the resolution of physical devices.pg 120
- an image that requires interpolation of pixels (scaling up) cannot be done w/o loss of image quality. pg 120
- DOWN SAMPLING - if an images resolution is higher than that of the output device, pixels must be discarded when the image is scaled down pg 121
- effects of downsampling..(creates a pradox)an image that has bee downsampled for display at a low resolution will often be better than that of an image whose resolution is equal to the display resolution. pg 121
- OVERSAMPLING - sampling an image (or any other signal) at a higher resolution than that at which it is ultimately displayed. pg 121
- disadvantage of high resolution is they contain more pixels and take longer to transfer over a network. pg 122
- RUN LENGTH ENCODING (RLE) - replacing a run of consecutive pixels of the same color by a single copy of the color value and a count of the number of pixels in the run. pg 123
- RLE has an important property; it is always possible to decompress run length endcoded data and retrieve an exact copy of the original data as it was before it was compressed. pg 123
- RLE is an example of lossless compression technique. pg 123
- compression - an operation performed on data to reduce the space required to represent it for reasons of economy or efficiency.Frequently used with bitmap images.
- lossy compression technique - information is discarded and can never be retrieved during compression. the result is only an approximation of the original and with each compression /decompression the quality will deteriorate. pg 124
- lossless compression can be allpied to any sort of data, its the only sort of compression that can be applied to data such as spreadsheet data,text, or binary executable programs. pg 125
- variable length codes In coding theory a variable-length code is a code which maps source symbols to a variable number of bits.-Variable-length codes can allow sources to be compressed and decompressed with zero error (lossless data compression) and still be read back symbol by symbol. With the right coding strategy an i.i.d. source may be compressed almost arbitrarily close to its entropy.
- dictionary based compression - works by constructing a table,or dictionary, into which are entered strings of bytes that are encountered in the input data; all occurences of a string are then replaced by a pointer into the dictionary. pg 124
- How does dictionary-based compression save space?
- Ideally entries should be long strings that occur frequently and the effectiveness depends on choosing strings to enter that replace them by their own codes.
- two techniques of construction a dictionary - (1) LZ77 (2) LZ78. a variation of LZ78 exist know as LZW and is one of the most widely used compression methods. pg 125
- LZW - basis of Unix compress and GIF. pg 125
- huffman coding- analyzes information and looks for common sequences of bits and replaces the most common with a smaller respresentation and less common with larger representation. picks out commonly seen sequences.
- image data can tolerate a certain amount of lost data so lossy compression can be used effectively. pg 125
- JPEG Compression - the most important lossy image compression technique. pg 125
- JPEG - best suited for for photographs and similar images which are characterized by fine detail. pg 125
- JPEG works by actually transforming an image into its frequency components. pg 126
- discrete cosine transform (DCT) takes an array of pixels and produces an array of coefficients, representing the amplitude of the frequency components in the image. pg 126
- It is impractical to imply DCT to an entire image at once, DCT images are divided into 8x8 pixel squares and each one is transformed separately, it is to computationally expensive to do so. pg 127
- JPEG compression- the number of quantization levels to be used for each frequency coefficient can be specified separately in a quantization matrix. pg 127
- JPEG compression- one highly useful feature, it is possible to control the degree of compression and thus the quality of the compressed image bi altering the values in the quantization matrix.
- compression artifacts - unwanted features in a compressed image, such as sharp edges come out blurred. pg 129
- What is one artifact of compressing that an image file can exhibit when using JPEG Compression? ANSWER: Since JPEG signs fewer quantization levels to higher frequencies sharp edges like those found in lettering and graphs get blurred.
- What is it about human vision that JPEG compression takes advantage of in order to compress a file? ANSWER: Humans can not distinguish high frequency changes in an image as well as they can detect low frequency changes. JPEG assigns less importance to the high frequency changes by assigning fewer quantization levels to them.
- zig zag sequence - the method in which coefficients are processed to maximize the length and runs of zeros
- filters and mask - pre-existing techniques used for altering photographic images. pg 130
- two broad reasons for manipulating an image - (1) correct deficiencies in an image. example, red eye in a photograph. (2) creating special effects. example, creating a glow around an object. pg 130
- Photoshop is the industry standard for image manipulation. pg 131
- bitmapp manipulation is primarily used for preparing images for print and are not relevant to multimedia pg 130
- multimedia use of bitmap manipulation is the changing of an images resolution or size, often images must be downsampled for a display or monitor. pg 130
Sections, Mask and Alpha Channels
- alpha channel - a greyscale mask used to specify different degrees of transparency in an image. (glossary)
- alpha channel - is like a stencil made out of material that can allow varying amounts of paint to pas through it, depending on the transparency value at each point. pg 133
- What's the Alpha Channel?
- The Alpha Channel is this nifty storage area in your image's file for non-visible picture elements. Think of it as your picture's own little private utility closet/filing cabinet. The .psp picture file has these extra goodies stored in its Alpha channel that can just travel around with the picture (unseen until you need them) as a part of the file. (fall semester study guide)
- mask - let you isolate and protect areas of an image as you apply color changes, filters, or other effects to the rest of the image.
Pixel Point Processing
- image procesing is performed by computing a new value for each pixel in an image pg 136
- simplest method to compute a new pixels value is solely based on its old value, w/o regard to any other pixel value pg 136
- how to compute new pixel value... p'=f(p) "f" is called the mapping function. Such functions perform pixel point procesing. pg 136
- most sophisticated pixel point processing is concerned with color correction and alteration pg 136
- crudest pixel adjustments made with brightness and contrast sliders pg 136
- brightness adjust the value of the pixel up or down uniformly and changes the intercept the line and the "Y" axis. pg 136
- contrast adjust the range of values , either enhancing or reucing the difference between the lightest and darkest areas of the image.adjusting the contrast alters the gradient of the line pg 136
- levels dialogue - allows you to move the endpoints of a linear mapping function individually, setting the black and white levels in the image. pg 136
- image histogram - a tool used to help set suitable adjustments (levels) to stretch or shrink the mapping function horizontally and vertically. pg 136
Pixel Group Processing
- pixel group processing - worksby computing each pixels new value as a fuction not just of its old value, but also of the value of neighboring pixels. pg 139
- when using a filter the array of weights is called the convolution mask and the set of pixels used in the computation is called the convolution kernel. pg 141
- Blurring is often used in retouching scans, and mitigating the effects of digial artefacts. pg 143
- low pass filter - When using a blurring filter, one way to think about the concept is to imagine that the edges have been softened by rubbing together the color value of the pixels (in the same way as you blur edges in a pastel drawing by rubbing them with your finger. An alternative view, based on the concepts of signal processing, is that this operation produces a smoothing effect on the spatial waveform of the image, by filtering out the high frequencies (music production). pg 143
- geometrical transformation - It's a process of scaling, translation, reflection, rotation and shearing pg 148
- applying geometrical transformations to bitmapped images is nt straightforward, every pixel has to be transformed and this will often require that the image be resampled. pg 148
- interpolation - a method used to reconstruct to an acceptable degree of accuracy. pg 149
- bilinear interpolation - uses values of four adjacent pixels. pg 150
- bicubic interpolation - the interpolation is based on cubic splines, that is the intermediate values are assumed ti lie along a Bezier curve connectiong the stored pixels , instead of a straight line, this process takes longer than others. pg 151
Additional Chapter Five Notes
- A 14-inch monitor provides a 640x480 pixels display.
- A 17-inch monitor provides 832x624 pixels display.
- Both 14-inch & 17-inch monitor display at 72 dpi
- A PAL frame is 768x576 pixels
- An NTSC frame is 640x480 pixels
- What is the difference between GIFS and PNGS?
- pngs can go between transparent and opaque while gifs can only go transparent or opaque
- red eye in photographs is caused by light reflecting off the eyes RETINAS.
CHAPTER SIX NOTES: COLOUR
Colour and Sceince
- color is a subjective sensation produced in the brain. pg 157
- spectral power distribution (SPD) a description of how the intensity of light from some particulsr source varies with wave length. pg 157
- wavelength of visible light lies roughly between 400nm and 700nm. pg 157
- A nanoMeter is 1 billionth of a meter...
- HINT: It’s a very short wave
- SPD's are to cumbersome to work with when we are secifying colors for use in computer graphics. pg157
- human eye contains two different sorts of receptor cells
- rods - which privide night vision and cannot distinguish color (only blk and wht)
- cones (3 different types) respond to different wavelengths of light(1 for red, 1 for green, 1 for blue)
- Tristimulus Theory - any color can be specified by just three values, giving the weight of each three values. pg 158
- Tristimulus Theory - each "CONE" responds to one RED, GREEN or BLUE light. pg 158
- Additive Primary Colors - Red,Green, Blue pg 158
- a pixel is made up of "3 dots" RGB, equalling 3 bytes. pg 158
- optical mixing of light emmited by three component dots of any pixel will make it look like a single pixel of the desired color. pg 158
- RBG color model - colors can be constructed out of red green blue given the the proportions of red, green, blue light which must be combined to make up light if the desired color. pg 158
- there is no real standard for red, green, and blue on computer displays, and the colors produced in response to any particular RGB value can vary between monitors. pg 159
- A color can be represented by three values, representation is written as (R,G,B) where R,G,B are the amounts of red green and blue light make a color.
- "AMOUNT" means the proportion of pure (saturated) light for that primary.
- Example; (100%,0%,0%)= pure saturated primary RED, (0%,100%,0%)=pure saturated primary GREEN, (0%,0%,100%)=pure saturated primary BLUE.
- (255,255,255)=WHITE, (0,0,0,)=BLACK
- Different shades of grey can be represented by the mixtures of RBG light with equal values. ex (34,34,34) OR (54,54,54), 2 different shades of grey.
Additive mixing -three values (#%,#%,#%) of amount of light of three primary colors which must be mixed to produce a specific color. pg 160
- the three #'s of RGB are not absolute values, it is only their relative values that matter. pg 160
- each value of RGB is 256 different values. 0 thru 255 = 8 bits each = 1 byte = 24 bits or 3 bytes total. pg 161
- 16.8 million different color combinations using 256 values each. pg160
- color depth - the number of bits used to hold a color value. pg 161
- 24 bit color - when 3 bytes are used, shorthand for 24 bit color depth. pg 161
- A SINGLE BIT ALLOWS US TO REPRESENT 2 DIFFERENT COLORS,(usually black and white) pg 161
- 4 bit color depth allows 16 different colors. pg 161
- Color Depth
- Usually expressed in bits
- One byte for each of the RGB => 24 bits
- Back to binary...
- 1 bit => 2^1 => 2 choices
- 2 bits => 2^2 => 4 choices
- 4 bits => 2^4 => 16 choices
- 8 bits => 2^8 => 256 choices
- greyscale images - represented by (RBG) color values r=g=b (34,34,34) pg 161
- RGB (25, 25, 25) is dark grey
- RGB (200, 200, 200) is light grey
- there can be a max of 256 different shades of gray
- some computer systems use 16 bits to hold color values
- 16 is not divisible by 3 (rgb)
- therefore one bit is unused or different # of bit are assigned to the three vaues.
- in this case red and blue use 5 bits and green is allocated 6 bits
- the human eye is more sensitive to green light than the other two primary colors this is the reason for the 6 bit allocation to green pg 162
- color depth terms - millions of colors (24bit), thousands of colors(16bit) 256 colors(8 bit) pg 162
- True color refers to 24 bit color pg 163
- Hi color refers to 16 bit color pg 163
- COLOR DEPTH IS A CRUCIAL FACTOR IN DETERMINING THE SIZE OF A BITMAPPED IMAGE. pg 163
- Any reduction in color depth has a 3-fold effect on the final image size
- A 100X100 RGB image
- at 24 bit color => 30,000 bytes uncompressed
- at 16 bit color => 20,000 bytes uncompressed
- 1 byte => 1/3 reduction of size
- Posterization- is when colors are distorted and details are lost. two similar colours are replaced by the same one, and banding and other visible artefacts may appear where gradations of color are replaced by sharp boundaries.
- two constraints on monitor colour depth
- an image file of 24 bit color may be considered too large for available disc space or it may take too long to transmit over a network.
- or that the video ram (vram) provided in the monitor is not sufficient to hold a full screen image at 3 bytes per pixel.
- palette- an index of 256 specific colors with each image.
- one way of thinking about index color is as a digital equivilant of painting by numbers.
- logical colors stored for logical pixels
- physical colors displayed as pysical pixels
- GIF supports 8 bit index color images
- JPEG images stored in 24 bit color must be reduced to 8 bit if necessary at the time of display
- dithering- done by grouping together pixels in an area to be colored and applying color to individual pixels in each group in a suitable pattern.
- this process is an extension of the use of half toning.
- dithering, at low resolutions may produce poor results it is better suited for high resolution work.
- web safe palette- a restricted set of 216 colors, the only palette you can rely on to be reproduced by web browsers on any system using 8 bit color.
Other Colour Models
- RBG color model is the most important means of representing color used in images for multimedia because it corresponds to the way in which color is produced on computer monitors and also how color is detected by scanners.
- C= cyan
- M= magenta
- Y= yellow
- K= black
- Complimentary Colors
- C= G+B=W-R
- M= R+B=W-G
- Y= R+G=W-B
- The color on the left is called complimentary color of the one on the extreme right.
- Example: Magenta is a complimentary color of green
- Examples of CMYK materials would be ink, paper and paint.
- CMYK is an Additive model (like RGB) and appropriate for use with light
- CMYK also provides the Subtractive model that is appropriate for use with ink and paint... things that absorb light'
- important point to grasp is that the light that is reflected from a color surface is not changed in color by the process of reflection.
- process colors -the four colors cyan, yellow, magenta and black in printing.
- CMYK colors are hard to manufacture because some unwanted colors are absorbed at the same time.
- Hue- the wavelength at which most of the energy of the light is concentrated (the dominate wavelength)
- there are 4 hues in the color spectrum, red, yellow, green, blue
- Isaac Newton identified seven hues, red, orange, yellow, green, blue, indigo, & violet of the rainbow.
- a pure hue can be more or less diluted by mixing it with white; the dominate hue remains the same but the presence of other hues make the color paler.
- a color's appearance will be modified by the intensity of light
- less light makes it appear darker
- the brightness of a color is a measure of how light or dark it is
- in terms of paint, a hue is a pure color
- adding white decreases its saturation producing a tint
- adding black decreases its brightness producing a tone
- a color saturation is measured by its purity
- HSV, HSB, HSL
- Hue, is the dominant frequency,expressed as degrees from red
- Saturation, how pale the hue is; adding white to paint reduces the saturation
- value. brightness, lightness. The color’s value, brightness or lightness;(adding black to paint reduces the brightness)
- You probably relate to Cyan to other blues you know (hue), how pale it is (saturation) and its brightness (intensity-V)
Colour Spaces based on colour differences
- to produce a measure of brightness from an RGB value, you must weigh the three componets separately
- Y= .2125R + .7154G + .0721B
- the quantity Y defined here is called luminance
Device-independent colour spaces
- RGB and CMYK are the most commonly used color models but they are both device dependent.
- device dependent means
- different monitors provide different red green and blue primaries
- different source of ink and paper produce different CMYK
Channels and Colour Correction
- channel- each of three gray scale images making up a color image
- the blue screen technique is commonly used in video and film making as well as constructing face images.
- listen to lecture for more discription
- Different devices use different phosphors
- 'Physical devices are not usually linear, they curve
- Red, Green and Blue phosphors each react differently and
- They do not react linearly... this means
- 10 times as much excitation will not produce 10 times as much emitted light
- In the end, it is all a compromise
- Phosphors and excitation circuitry age