Very well stated, Hector.
In my experience, where up to 3,000 tests per day are performed, over a period of several weeks, it would be impossible to do it mechanically, besides being prohibitively expensive.
The human nose is still best for picking up nuances that would be missed or misinterpreted by a machine.
Take Care,
George Berkompas
----- Original Message -----
From: "hmlupin@libero.it" <hmlupin@libero.it>
Date: Saturday, October 13, 2007 3:24 am
Subject: Re:electronic noses
To: doctoredwin2003 <doctoredwin2003@yahoo.com>
Cc: seafood <seafood@ucdavis.edu>
> Dear Edwin,
>
> Even if some devices have been developed on the line you mention,
> they definitely will not be cheap (not to say reliable enough) to
> utilize with fresh (and frozen) Nile Perch fillets. As you mention in
> your note you have an integrated problem, texture + colour + odour.
>
> The machine you want should be able to touch, to see (in colour
> according with the wavelength radiation humans see) and to smell (if
> not as a human at least some key volatile compounds). Not only should
> that produce a single output answer, which integrates (in some way)
> all the single evaluations, which means a sort of computer. I would be
> tempted to add that the computer should have a "self learning"
> programme that allow it to gain experience from each lecture.
>
> At this point we can ask ourselves why to design such a machine (that
> is theoretically possible) if humans can perform all that much more
> easily, cheaply and despite unavoidable errors, surely, at the end,
> with more reliability than machines.
>
> In addition humans eat fish and machines don't. This is very
> important because we are discussing here about aspects of (edible)
> quality of interest to consumers. There are many aspects that go into
> sensory evaluations that escape sensory score tables.
>
> To perform sensory analysis the key point is to be well trained, but
> this is not difficult, if somebody could help you to take the first
> steps and think how to do. The key point to remember is that sensory
> analysis is not apply blindly score tables, but to think about what
> are you doing (the "computer" that integrates all and learn something
> more from each lecture).
>
> Fortunately in your country you have very good and gifted fish
> technologists to whom to ask for assistance. They were working on the
> problem you have today more than 20 years ago. Start with Frank
> Nanyaro (from the Fish Inspection Service) that surely will tell you
> what to do. But of course there are other people like John Makene,
> Michael Mlay, Thomas Maembe and their disciples that surely can help
> you. You can tell them that I indicated you to do so. As a general
> recommendation look at your country first, give credit and trust your
> own people.
>
> Kind regards.
>
> Hector M. Lupin
>
>
>
> ---------- Initial Header -----------
>
> >From : owner-seafood@ucdavis.edu
> To : "Seafood research and extension information exchange" seafood@ucdavis.edu
> Cc :
> Date : Mon, 8 Oct 2007 05:43:46 -0700 (PDT)
> Subject : electronic noses
>
>
>
>
>
>
>
> > Hi all,
> >
> > could anyone update me if an electronic instrument that can be
> used to substitute manual
> >
> > organoleptic selection of fish fillets is available in the
> market.We have conducted trials with
> >
> > the torrymeter but it indicates poor/good texture only ,it does
> not help with colour or odour.
> >
> > We process nile perch fillets from lake victoria.
> >
> > Thanks and Best Regards
> >
> > Edwin Okong'o
> > Mwanza Fishing Industries Ltd'
> > P.O.Box 348
> > Mwanza Tanzania
> >
> >
> > ---------------------------------
> > Shape Yahoo! in your own image. Join our Network Research Panel today!
>
>
> ------------------------------------------------------
> Leggi GRATIS le tue mail con il telefonino i-mode™ di Wind
>
>
This archive was generated by hypermail 2b29 : Tue Oct 16 2007 - 08:33:27 PDT