Re: A challenge: What is an edge?


Subject: Re: A challenge: What is an edge?
From: Breezy (weiz@spam-block.sympac.com.au)
Date: Tue Apr 18 2000 - 07:35:01 MET DST


I like fuzzy sets because they easily accept, in fact need, real life view
points for definition and in doing so fit so nicely many real life
circumstance.

the members of a tall fuzzy set have as much to do with the system that
defines tall as they do to the members of the set.

 A group of short people would likely find six feet high individuals a
tallish proposition, where as a bunch of basket ball types would likely find
these folk short.

To edge. If the cylinder is hot and its background cold but both are the
same color then now edge, clearly depends on the fuzzy measure and is likely
to be very different propositions depending on view.

I would be inclined to define a fuzzy edge from the stand point of the view
instructing the system wanting the definition.

I could imagine that a poll of onlookers (the local knowledge about the
edge) could score (for a visual learning device) their degree of certainty
as to a visual edge and a composite of those scores could then render an
edge concept. At points definite object and at others definite background
and then at the juncture, to some degree of membership certainty, it will be
both. A real uncertain fuzzy edge (with very close matching colors and
textures of object and background) will have a larger fuzzy range. A crisp
edge (black & white without shadow or shine) will be obvious right down to a
definitive point, but it is defined by the view point, or what could be
thought of as the on hand expert.

Turn the lights right off and there is no edge, no object and no view.

Why some amorphous facet ? Is not an edge a great way to use the already
powerful and perfectly simple fuzzy set theory.

Rob W

Isaac wrote in message <9wdK4.743$4_4.313271@news.pacbell.net>...
>I may be green in the field, but wouldn't an edge have two main components:
>1. A physical gradient of some perceivable quality (local)
>2. A psychological gradient, or logical organization boundary (global)
>
>I would think that the 3 categories you suggest are really a subset of a
>continuos rage of confidence levels in the above 2 items. That is, cisp=
>high gradient, fuzzy= medium, and amorphous = low. Most methods rely on
>item 1, but in the "amorphous" case, it seems, item 2, above, would be
>required to increase confidence levels, or member function lines, by
>admitting some underlying structure that global edges offer to
indeterminate
>local ones.
>
>Hope I'm not off base here.
>
>Appreciate your thoughts as well,
>Isaac-
>
>
>Michelle T. Lin <michlin@cs.berkeley.edu> wrote in message
>news:200004040013.RAA05336@typhon.CS.Berkeley.EDU...
>> *********************************************************************
>> Berkeley Initiative in Soft Computing (BISC)
>> *********************************************************************
>>
>>
>> To: BISC Group
>> From: L. A. Zadeh <zadeh@cs.berkeley.edu>
>>

############################################################################
This message was posted through the fuzzy mailing list.
(1) To subscribe to this mailing list, send a message body of
"SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at
(2) To unsubscribe from this mailing list, send a message body of
"UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com"
to listproc@dbai.tuwien.ac.at
(3) To reach the human who maintains the list, send mail to
fuzzy-owner@dbai.tuwien.ac.at
(4) WWW access and other information on Fuzzy Sets and Logic see
http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info
(5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html



This archive was generated by hypermail 2b25 : Tue Apr 18 2000 - 07:55:33 MET DST