© The Institution of Electrical Engineers
Pattern recognition has usually been regarded as a two-part process requiring both measurement and classification. It has also been accepted by many workers that the classifier can best be designed using an automatic learning process. However it is well known that many of the proposed learning schemes sometimes fail to realise the most efficient use of the available storage. This difficulty usually arises whenever the initial classifier was a poor approximation to the teacher. The design of the classifier using such learning techniques would clearly be more successful if a method could be found of providing a good starting state for the learning machine. The letter defines an algorithm and examines its performance in this task of initialising a learning machine.
References
-
-
1)
-
B.G. Batchelor
.
(1969)
A hybrid pattern classifier for computer input, Computer science and technology.
-
2)
-
Batchelor, B.G.: `Learning machines for pattern recognition', 1968, Ph.D. thesis, University of Southampton.
-
3)
-
J.R. Parks
.
Automatic recognition of low quality printed characters using analogue techniques.
Radio Electron. Engrs.
,
67 -
80
-
4)
-
B.G. Batchelor ,
B.R. Wilkins
.
Adaptive discriminant functions.
IEE Conf. Publ.
-
5)
-
G.S. Sebestyn
.
(1962)
, Decision making processes in pattern recognition.
-
6)
-
N.J. Nillson
.
(1965)
, Learning machines.
-
7)
-
Parks, J.R.: `An analogue technique for the recognition of low quality printed characters', 17, National Physical Laboratory Autonomics Division Report, 1964.
http://iet.metastore.ingenta.com/content/journals/10.1049/el_19690366
Related content
content/journals/10.1049/el_19690366
pub_keyword,iet_inspecKeyword,pub_concept
6
6