system architecture the integration of processing components and knowledge

50
System Architecture The Integration of Processing Components and Knowledge

Upload: nancy-oneal

Post on 05-Jan-2016

221 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: System Architecture The Integration of Processing Components and Knowledge

System Architecture

The Integration of Processing Components and Knowledge

Page 2: System Architecture The Integration of Processing Components and Knowledge

Introduction

• So far– Presented methods of achieving goals

• Integration of methods?– Controlling execution– Incorporating knowledge

Page 3: System Architecture The Integration of Processing Components and Knowledge

3

Knowledge

“The fact of knowing a thing, state, person; A state of being aware or informed; Consciousness”.

Shorter Oxford English Dictionary (1973)

Page 4: System Architecture The Integration of Processing Components and Knowledge

4

Knowledge and Intelligence

Knowledge

Intelligence Cognition

Language; Creativity; Planning, Thinking,

Computation

Page 5: System Architecture The Integration of Processing Components and Knowledge

5

Knowledge and Intelligence

•Knowledge is acquired and disseminated by intelligent and cognate beings. The terms knowledge, cognition and intelligence are used interchangeably.

•And there is a good reason for this: Various cognitive processes help in converting information and stimuli into knowledge. Knowledgeable beings then act intelligently because of their greater awareness.

Page 6: System Architecture The Integration of Processing Components and Knowledge

What knowledge?

• What do algorithms achieve?

• What is known about the problem being solved?

• Relationship between problem and algorithm?

Page 7: System Architecture The Integration of Processing Components and Knowledge

Knowledge representation

Each knowledge representation scheme must include:• Strategic knowledge about the problem being

investigated and potential solutions• Algorithmic knowledge about how potential

solutions translate into computer vision algorithms and how the algorithms interact (about the control)

• Data knowledge about information derived from the image and how this can contribute to the solution

Page 8: System Architecture The Integration of Processing Components and Knowledge

Knowledge representation

• Implied

• Feature vectors

• Relational structures

• Hierarchical structures

• Rules

• Frames

Page 9: System Architecture The Integration of Processing Components and Knowledge

Implied knowledge

• Knowledge encoded in software

• Usually inflexible in– Execution– Reuse

• Simple to design and implement

• Systems often unreliable

Page 10: System Architecture The Integration of Processing Components and Knowledge

Feature vectors

• As seen in statistical representations

• Vector elements can be– Numerical– Symbolic coded numerically

Page 11: System Architecture The Integration of Processing Components and Knowledge

Example:

strokes 3

loops 1

w-h ratio 1

A Nstrokes 3

loops 0

w-h ratio 1

Page 12: System Architecture The Integration of Processing Components and Knowledge

Relational structures

• Encodes relationships between– Objects– Parts of objects

• Can become unwieldy for – Large scenes– Complex objects

Page 13: System Architecture The Integration of Processing Components and Knowledge

Relational structures

supporting supportingblock

blockblockadjacent

adjacent

above above

Page 14: System Architecture The Integration of Processing Components and Knowledge

Follow natural division of

Hierarchical structures

scene

objects

parts of object

Page 15: System Architecture The Integration of Processing Components and Knowledge

Example:scene

roadway building grassland

grass treeroad junction

edges

Page 16: System Architecture The Integration of Processing Components and Knowledge

Uses

• Structure defines possible appearance of objects

• Structure guides processing

• Representing a processing result

Page 17: System Architecture The Integration of Processing Components and Knowledge

Rule-Based System

Database RulebaseInferenceengine

Page 18: System Architecture The Integration of Processing Components and Knowledge

Rules

• Rules code quanta of knowledge

• Interpretation– Forwards– Backwards

<antecedent> <action>

<two antiparallel lines> <road>

Page 19: System Architecture The Integration of Processing Components and Knowledge

Forward chaining

• If <antecedent> is TRUE

• Execute <action>

• Antecedent will be a test on some data

• Action might modify the data

• Suitable for low level processing

Page 20: System Architecture The Integration of Processing Components and Knowledge

Backward chaining

• Action is some goal to achieve

• Antecedent defines how it should be achieved

• Suitable for high level processing– Guides focus of system

Page 21: System Architecture The Integration of Processing Components and Knowledge

FramesA “data-structure for representing a

stereotyped situation”

Slot(attribute) Filler

(value: atomic, link to another frame, default or empty, call to a function to fill the slot)

Page 22: System Architecture The Integration of Processing Components and Knowledge

Methods of control

• How to control how the system’s knowledge is used.– Hierarchical

Page 23: System Architecture The Integration of Processing Components and Knowledge

Hierarchical control

• “Algorithm” defines control

• Two extreme variants– Bottom-up– Top-down

Page 24: System Architecture The Integration of Processing Components and Knowledge

Bottom-up control

Objectrecognition

Extracted features,Attributes,

Relationships

Image

Decision making

Feature extraction

Page 25: System Architecture The Integration of Processing Components and Knowledge

Top-down control

Hypothesisedobject

Predicted features,Attributes,

Relationships

Features in image thatSupport or refute the

hypothesis

Prediction

Directed feature extraction

Page 26: System Architecture The Integration of Processing Components and Knowledge

Critique

• Inflexible methods

• Errors propagate

• Hybrid control– Can make predictions– Verify– Modify predictions

Page 27: System Architecture The Integration of Processing Components and Knowledge

Hybrid control

Objectrecognition

Image

Decision making

Feature extraction

Extracted features,Attributes,

Relationships

Predicted features,Attributes,

Relationships

Prediction

Page 28: System Architecture The Integration of Processing Components and Knowledge

Uncertainty Reasoning

• Bayesian methods– Define a belief network– A tree structure

• Reflects evidential support of a fact

F1

F2 F3

Page 29: System Architecture The Integration of Processing Components and Knowledge

Dempster-Shafer

• Bayesian theory has confidence in belief only

• No measure of disbelief

• D-S attempts to define this

Page 30: System Architecture The Integration of Processing Components and Knowledge

Dempster-Shafer

• Bayesian reasoning allows us to state our belief in a hypothesis and our belief in that same hypothesis when some new data are received.

• Dempster-Shafer theory (D-S) also provides an assessment of belief in some hypothesis which can be modified in the light of new data.

• Unlike Bayesian reasoning, D-S takes into account the fact that it may not be possible to assign a belief to every hypothesis set.

Page 31: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• Mrs Jones has a carton of cream delivered along with the milk every early morning on some days of the week. On most mornings following delivery of the cream, the carton is found open and the content is gone. Mrs Jones believes that the culprit is one of the three animals that stalk the area. One animal is a dog, the other a cat and the third a fox. Occasionally a neighbour will catch sight of the thief in the act, but the delivery is before daylight and no neighbour has been certain about their sighting.

Page 32: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

There are three suspects:– Dog –d– Cat – c– Fox – f

and each suspect represents a hypothesis. Only a single animal is responsible for the theft.

Page 33: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• The set of hypotheses is called the frame of discernment Θ. In this example:

• Θ={d,c,f}• The thief is either the dog or the cat or the fox. • DS is not limited to assigning a belief to only dog,

cat or fox but can assign beliefs to any element that is a member of the power set of Θ.

• The belief in an element, x, is referred to as a probability mass denoted, m(x).

Page 34: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• The power set of Θ is the set of all subsets of Θ and is denoted by 2 Θ .

• 2 Θ={Φ, {d}, {c}, {f}, {d,c}, {d,f}, {c,f}, Θ}• Where Φ denote the empty set. • The power set expresses all possibilities. For

example, {d} is the hypothesis that the dog takes the cream and {d,f} is the hypothesis that the culprit is either the dog or the fox.

Page 35: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• There are restrictions on the values of m(x):

• Which state that the total mass must sum to 1 and that the empty set is not possible (the closed world assumption which means that no animal other than the dog , fox or cat is stealing the cream).

• Any subset x that has a non-zero value for m(x) is called a focal element.

1)(2

x

xm 0)( m

Page 36: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• Suppose neighbour 1 states that she believes it is either the dog or cat with probability 0.8. So m({d,c}) = 0.8.

• The probability must sum to 1 and so 0.2 has to assigned somehow to the other hypotheses sets. The best we can do without any other information is to assign it to the whole frame of discernment m({d, c, f}) = 0.2.

Page 37: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• On the following night, another neighbour spots the thief and states that she believes that it was either the cat or fox with probability 0.7.

• How should these new data be combined with the original data? D-S theory states that the original mass is combined with the new mass according to the rule

-eqn (1)

BAC

BmAmCm )()()(

Page 38: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• A is the set of focal elements identified by neighbour 1 and B those by neighbour 2. This equation states that there is a set C of focal elements formed by the intersection of the sets in A and B and the mass assigned to an element in C is the product of the intersection masses. The result of applying eqn (1) is given in table 1

Page 39: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

Neighbour 2

m({c,f}) =0.7 m({d,c,f})=0.3

Neighbour 1 m({d,c})=0.8 m({c})=0.56 m({d,c})=0.24

m({d,c,f})=0.2 m({c,f}) =0.14 m(d,c,f})=0.06

Table 1. The probability masses from neighbours 1 and 2 are combined

Page 40: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• We shall use the notation mn to indicate the evidence has been encountered at step n. The first step was from neighbour 1 and the second from neighbour 2, which are combined to give a new belief at step 3. So:

m3({c})=0.56m3({d,c})=0.24m3({c,f}) =0.14m3(d,c,f})=0.06

Page 41: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• Two probability measures are provided which assess the belief (Bel) and plausibility (Pl) of any set of hypotheses:

-eqn (2)

-eqn (3)

 

AB

BmABel )()(

AB

BmAPl )()(

Page 42: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• These two measures represent lower and upper bounds on the belief in a set of hypotheses. So the belief in the cat being the culprit is the sum of the masses where the set of hypotheses is a subset of {c}, which in this case is simply:

• Bel({c})=0.56• The plausibility is the sum of all masses that

contain cat as a member:• Pl ({c}) = 0.56+0.24+0.14+0.06= 1.0

Page 43: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• The belief and plausibility in the dog and fox are:

• Bel({d}) = 0

• Pl({d}) = 0.24+0.06 = 0.3

• Bel({f}) =0

• Pl ({f}) 0.14+0.06=0.2

Page 44: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• The belief and plausibility in it being either the dog or cat are:

• Bel ({d,c}) = 0.56+0.24=0.8

• Pl({d,c}) = 0.56+0.24+0.14+0.06=1.0

Page 45: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

Neighbour 3

m4({f})=0.6 m4=({d,c,f})=0.4

m3({c})=0.56 null m5({c})=0.224

m3({d,c})=0.24 null m5({d,c})=0.096

Existing focal elements

m3({c,f}) =0.14 m5({f})=0.084 m5({d,f})=0.056

m3(d,c,f})=0.06 m5({f})=0.036 m5({d,c,f})=0.024

Table 2. Combining evidence from neighbour 3 with the evidence derived from combining sightings of neighbours 1 and 2

Page 46: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• Table 2 is problematic because there are two null entries that indicate an empty intersection between the existing focal elements and the new evidence. In other words, the empty set has a mass which violates the earlier condition that is not possible to have belief in something outside of the sets of hypotheses. The suggested way around this problem is to normalise the entries using the following equation

BA

nnCBA

nn

n

BmAm

BmAmCm

)()(

)()()(

1

1

2

Page 47: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

• For our example, this equation suggests that we should divide each new focal element by the sum of all focal elements that do not have a null entry. All we are doing is ensuring that the null entries have a mass of zero and that all other new focal elements sum to 1. The denominator is:

• 0.084+ 0.036 + 0.224 + 0.096 + 0.056 + 0.024 = 0.52• Each newly calculated focal element in table 2 is now

updated by dividing by 0.52. The updated values are given in table 3. The final beliefs and plausibilities for each set of hypotheses after all three neighbours have given evidence are list in table 4

Page 48: System Architecture The Integration of Processing Components and Knowledge

D-S scenarioNeighbour 3

m4({f})=0.6 m4=({d,c,f})=0.4

m3({c})=0.56 null m5({c})=0.431

m3({d,c})=0.24 null m5({d,c})=0.185

Existing focal elements

m3({c,f}) =0.14 m5({f})=0.162 m5({d,f})=0.108

m3(d,c,f})=0.06 m5({f})=0.069 m5({d,c,f})=0.046

Table 3

Page 49: System Architecture The Integration of Processing Components and Knowledge

D-S scenario

Belief Plausibility

{d} 0 0.231

{c} 0.431 0.770

{f} 0.231 0.385

{d,c} 0.616 0.770

{d,f} 0.231 0.570

{c,f} 0.770 1.0

{d,c,f} 1.0 1.0

Table 4

Page 50: System Architecture The Integration of Processing Components and Knowledge

Summary

• Intelligent (vision) systems– Knowledge representation– Control strategies– reasoning