Newsgroups: comp.graphics.algorithms,comp.lang.c,de.comp.lang.c,sci.image.processing,comp.graphics.algorithms
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.mathworks.com!newshost.marcam.com!usc!howland.reston.ans.net!xlink.net!news.PFM-Mainz.DE!mainz.rhein-main.de!marvin!micha
From: micha@marvin.rhein-main.de (Michael Metzger)
Subject: Q: Histogram evaluation
Keywords: histogram, extremas, smoothing
Message-ID: <1995Feb12.165425.17361@marvin.rhein-main.de>
Summary: Looking for a package to get all sorts of information out of a histogram.
Date: Sun, 12 Feb 1995 16:54:25 GMT
X-Newsreader: TIN [version 1.2 PL2]
Lines: 36
Xref: glinda.oz.cs.cmu.edu comp.graphics.algorithms:13189 comp.lang.c:127167 sci.image.processing:12591

Hello out there!

I am currently involved in document-analasys techniques and
came across alot of decision-making on the basis of histograms.
What I not came across of ist a technique of getting the infos
out of the histograms the decisions are based on. So e.g. the
number of relevant extremas (not too close together, not too broad,
just right as you would judge it by looking at).

The statistical means of accomplishing these tasks are more or less
loosely connected in my mind, just enough to have an idea, of how
this histgram-toolbox should look like. But since I have rather harsh
time constraints currently I can't do it myself.

Who knows a package like this? Preferrably in ANSI-C, having a datastructure
holding the histogram-data and some operations on it performing smoothing,
getting the n most important maximas/minimas, get the nth derivative, perhaps
even doing some distribution-tests.

It would be of very much help indeed.

Thanks for reading, and thinking about it.

Regards
Michael

--
Michael Metzger
micha@marvin.rhein-main.de
+49-6131-834186
========================== end ==========================================

-- 
Michael Metzger                      tel: +49-6131-834186
                        email: micha@marvin.rhein-main.de

