: is anything that has specific descriptions in mathematical form.*Model*

operation of simplifying subsequent operations without losing relative information.*Preprocessing:*

operation of isolating image objects from background.*Segmentation:*

space containing all pattern features.*Feature space:*

is a vector from feature space that contains the actual values of current model.*Feature vector*

operation of reducing the data by measuring certain features or properties.*Feature Extraction:*

samples used to extract some information about domain of problem.*Training Samples:*

theory of making a decision rule in order to minimize cost.*Decision Theory:*

a boundary that distinguishes classes decision regions.*Decision Boundary:*

patterns that were not included in the training samples.*Novel Patterns:*

classifier ability to classify novel patterns in right class.*Generalization:*

technique used to resolve problem of insufficient training data by having a model of how each pattern was produced.*Analysis by Synthesis:*

techniques to process images for enhancements and other purposes.*Image Processing:*

technique used to discover representative features of a group of patterns.*Associative Memory:*

area of finding some functional description of data in order to predict new value.*Regression:*

area of inferring the function for intermediate ranges of input.*Interpolation:*

the problem of estimating the density (probability) that a member of a certain category will be found to have particular features.*Density Estimation:*

*Pattern Recognition System:*

is the study of part/whole relationships in order to make the classifier categorize inputs as “make sense” rather than matching but not too much.*Mereology:*

features that are invariant to irrelevant transformations of the input*Invariant (distinguishing) Feature:*

concept of hiding part of an object by other part*Occlusion:*

operation of selecting most valuable features from a larger set of candidate features*Feature Selection:*

any property of the sensed pattern which is not duo to the underlying model but instead to randomness in the world or the sensors*Noise:*

percentage of new patterns that are assigned to the wrong category*Error Rate:*

total expected cost*Risk (Conditional Risk):*

, idea about having several classifiers used to categorize one sample. Here how can we resolve the problem of disagreeing classifying?*Evidence pooling*

multiple classifiers working on different aspects of the input*Multiple Classifier:*

:*Design of Classification System*

: situation where the classifier is tuned on training samples and is unable to classify novel pattern correctly.*Overfitting*

: learning technique where a teacher provides a category label or cost for each pattern in training set.*Supervised Learning*

: learning technique where the system itself makes “natural groupings” of input patterns.*Unsupervised Learning (Clustering)*

Calculating a tentative value for each category label to improve classification.*Reinforcement Learning:*

**Chapter # 2
**

: class that the current sample belongs to.*State of Nature*

probability that the next sample will belongs to a specific class.*Prior:*

rule that decides for which class the current sample belongs to.*Decision Rule:*

: the probability of the state if nature being belongs to a specific class given that feature value x has been measured.*Posterior*

scaling factor that guarantees that the posterior probabilities sum to one*Evidence Factor:*

function that states how costly each action is, and is used to convert a probability determination into a decision.*Loss Function:*

minimum overall risk*Bayes Risk:*

: loss function where each action is assigned to its category (i=j)*Zero-One (symmetrical) Loss*

: region for each class on the histogram*Decision Region*

classifier that places a pattern in one of only two categories*Dichotomizer:*

classifier that places a pattern in more than two categories*Polychtomizer:*

a linear function that is able to discriminate classes on the histogram*Linear Discriminate Function:*

assigning x to the nearest mean*Template Matching:*