Syntactic pattern recognition: Difference between revisions

Revision as of 13:24, 21 January 2010

Revision as of 14:14, 21 January 2010

Line 1:

'''Syntactic pattern recognition''' or '''structural pattern recognition''' is a form of [[pattern recognition]], in which each object can be represented by a variable-[[cardinality]] set of symbolic, ([[nominal) data|nominal]] features. This allows for representing pattern structures, taking into account more complex interrelationships between attributes than is possible in the case of flat, numerical [[feature vector]]s of fixed dimensionality, that are used in [[statistical classification]].

Syntactic pattern recognition can be used instead of statistical pattern recognition if there is clear structure in the patterns. One way to present such structure is by means of a [[String (computer science)|strings]] of symbols from a [[formal language]]. In this case the differences in the structures of the classes are encoded as different [[formal grammar|grammars]].

Line 5:

An example of this would be diagnosis of the [[heart]] with [[Electrocardiogram|ECG]] measurements. ECG [[waveform]]s can be approximated with diagonal and vertical line segments. If normal and unhealthy waveforms can be described as formal grammars, measured ECG signal can be classified as healthy or unhealthy by first describing it in term of the basic line segments and then trying to parse the descriptions according to the grammars.Another example is [[tessellation]] of tiling patterns.

AnotherA second way to represent relations are [[Graph (mathematics)|graphs]], where nodes are connected if corresponding subpatterns are related. An item can be labeled as belonging to a class if its graph representation is [[isomorphic]] with prototype graphs of the class.

Typically, patterns are constructed from simpler subpatternssub patterns in a hierarchical fashion. This helps in dividing the recognition task into easier subtask of first identifying subpatternssub patterns and only then the actual patterns.

Structural methods provide description of items, which may useful on its own right. For example, syntactic pattern recognition can be used to find out what object are present in an image. Furthermore, structural methods are strong in finding a '''correspondence mapping''' between two images of an object. Under natural conditions, corresponding features will be in different positions in the two images, due to camera-attitude and perspective, as in [[face recognition]].

== See also==

Line 16:

== References ==

{{cite book | last = Schalkoff | first = Robert | title = Pattern recognition - statistical, structural and neural approaches | publisher = John Wiley & sons | year = 1992 | isbn = 0-471-55238-0 }}

{{cite book | last= Bunke | first = Horst | title = Structural and syntactic pattern recognition, Chen, Pau & Wang (Eds.) Handbook of pattern recognition & computer vision | publisher = World Scientific | pages = 163 - 209 | year = 1993 | ISBN = 981-02-1136-8 }}

[[Category:Classification algorithms]]