R help - Aug 2004 - image recognition in R

I have some images of bugs (insects) with many bugs in each image.
I want to count the number of bugs and to have an estimate of the area 
of each one.
I've tried searching for an R package to do so with no success. Is this 
a task that I should pursue doing in R or should I restrict myself to 
specific image analysis software (e.g. ImageJ)?.

The reason I consider R would be a good choice is because then It would 
probably be possible to use R statistical power to do pattern 
recognition on each bug's image to try to identify each bug (Family, 
genus or species).
Is anybody working in this direction?
Thanks,
Angel

Hello,
You should use ImageJ to binarize the picture and then extract features
(measurements) for each blob (individual). Save results as a text file and
import it in R (read.table). There you have plenty of packages to analyze
and classify your data. Things to try range from classical lda, to learning
vector quantization or neural nets (bundle VR), support vector machine
(package e1071) bagging (package ipred) and random forest (package
randomForest), among others. 

Otherwise, I work on a similar subject: I study methods for (semi)-automatic
recognition of plancton based on the analysis of digitized pictures. A
reference: Grosjean, Ph., M. Picheral, C. Warembourg & G. Gorsky, 2004.
Enumeration, measurement and identification of net zooplankton samples using
the ZOOSCAN digital imaging system. ICES J. Mar. Sci., 61:518-525. If you
are interested, contact me on my private email box.
Best,

Philippe Grosjean

..............................................<??}))><........
 ) ) ) ) )
( ( ( ( (    Prof. Philippe Grosjean
 ) ) ) ) )
( ( ( ( (    Numerical Ecology of Aquatic Systems
 ) ) ) ) )   Mons-Hainaut University, Pentagone
( ( ( ( (    Academie Universitaire Wallonie-Bruxelles
 ) ) ) ) )   8, av du Champ de Mars, 7000 Mons, Belgium  
( ( ( ( (       
 ) ) ) ) )   phone: + 32.65.37.34.97, fax: + 32.65.37.33.12
( ( ( ( (    email: Philippe.Grosjean at umh.ac.be
 ) ) ) ) )      
( ( ( ( (    SciViews coordinator (http://www.sciviews.org)
 ) ) ) ) )
..............................................................

> -----Original Message-----
> From: r-help-bounces at stat.math.ethz.ch 
> [mailto:r-help-bounces at stat.math.ethz.ch] On Behalf Of Angel Lopez
> Sent: Wednesday, August 25, 2004 12:05 PM
> To: r-help at stat.math.ethz.ch
> Subject: [R] image recognition in R
> 
> I have some images of bugs (insects) with many bugs in each image.
> I want to count the number of bugs and to have an estimate of 
> the area of each one.
> I've tried searching for an R package to do so with no 
> success. Is this a task that I should pursue doing in R or 
> should I restrict myself to specific image analysis software 
> (e.g. ImageJ)?.
> 
> The reason I consider R would be a good choice is because 
> then It would probably be possible to use R statistical power 
> to do pattern recognition on each bug's image to try to 
> identify each bug (Family, genus or species).
> Is anybody working in this direction?
> Thanks,
> Angel
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! 
> http://www.R-project.org/posting-guide.html
> 
>

Angel Lopez writes:
> I have some images of bugs (insects) with many bugs in each image.
> I want to count the number of bugs and to have an estimate of the area 
> of each one.
See also the book
    A. Lawson & D. Denison
    Spatial cluster modelling
    Chapman & Hall
where chapters 4-6 describe different statistical strategies for recognising
objects in an image. The authors of these three chapters 
each have some code, but it's probably not wrapped up in an R package yet.
And also 
  A.J. Baddeley and M.N.M. van Lieshout.
  Stochastic geometry models in high-level vision.
  In K.V. Mardia (ed.) Statistics and Images, volume 1,
  Abingdon: Carfax, 1993. pp. 233--258.