IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP


Luciano da Fontoura Costa luciano at OLIVE.IFQSC.SC.USP.BR
Fri Nov 10 19:17:13 EST 1995

10th Nove 95

Dear Colleagues:

I would like to inform that the papers below were published
recently.   Reprints can be obtained by sending a post-card
to the address below:

Luciano da F. Costa
Cybernetic Vision Research Group
IFSC - University of Sao Paulo
Caixa Postal 369,
13560-970    Sao Carlos, SP


Computer vision based morphometric characterization of neural
cells - L. da F. Costa, Review of Scientific Instruments,
66(7):3770-3773, July 95.

ABSTRACT:  This article reports on the application of computer
vision techniques, implemented on an IBM PC-compatible machine,
as a means of obtaining morphometric characterization of neural
cells.  The features considered include projection polar histograms,
orientation polar histograms, and the determination of the convex
hull of denritic (axonal) arborizations.  As will be discussed in
this work, the Hough transform technique has proven to be of great
help for the determination of all the above-mentioned parameters.
Two actual application examples respective to the analysis of retinal
ganglion cells are also presented and discussed.


Towards real-time detection of discrete straight features with a
hybrid technique based on preliminary curve segmentation and zoomed-
adaptive parametric mapping - L. da F. Costa, Real-Time Imaging,
1:203-213, 1995.

ABSTRACT:  A technique capable of high performance, real-time
detection of discrete straight line segments is described that
allows major improvements to traditional alternative techniques
such as the standard Hough transform.  Such improved features result
from the following three underlying principles: (a) the adoption of
a precise mathematical-theoretical characterization of straight lines
and their mapping into parametric spaces; (b) the incorporation of
adaptive parametric mapping that allows full congruency in the repre-
sentation of discrete straight lines; and (c) the inclusion of a pre-
processing stage that performs preliminary curve segmentation thus
constraining the parameter space to be further investigated, which is
done in zoomed fashion.  The high performance of the proposed technique
has been corroborated, comparatively to three alternative techniques
including the Hough transform, through a formal and quantitative
performance assessment that is also discussed.


The box-counting fractal dimension: Does it provide an accurate
subsidy for experimental shape characterization?  If so, how to use it?
by R.C. Coelho and L. da F. Costa.  SIBGRAPI 1995.

ABSTRACT:  This paper reports an experimental assessment of how
appropriate and accurate the self-similarity or box-counting fractal
dimension is for practical applications to the analysis and
characterization of natural objects.  The effects of the partial
fractality of real objects as well as the limited representation allowed
by typical spatially quantized images are investigated through a series
of experimental measurements respectively to two Koch curves.  Practical
guidelines are established and discussed that can hel in obtaining more
precise and meaningful experimental results.  These guidelines are
illustrated with respect to images of synthetic neural cells.


Book Review:  Dialogues on Perception (B. Julesz), by L. da F. Costa,
Real-Time Imaging, 1995.


Book Review: Artificial Neural Networks for Speech and Vision
(R. J. Mammone), by J. F. Fontanari and L. da F. Costa, IJNS, 1995.


Book review: The Computational Brain (Churchland and Sejnowski),
by L. da F. Costa.


Book review: The Cognitive Brain (A. Trehub), by L. da F. Costa,
Neurocomputing, 8:223-228, 1995.





Luciano da Fontoura Costa
Cybernetic Vision Research Group

IFSC - Universidade de Sao Paulo
Caixa Postal 369, Sao Carlos, SP

FAX:     +55 (162) 71 3616
e-mail:  Luciano at ifqsc.sc.usp.br
WWW:     http://scorpions.ifqsc.sc.usp.br/ifsc/ffi/grupos/instrum/visao/


    The Definition of an Upgrade: Take old bugs out, put new ones in.




More information about the Neur-sci mailing list

Send comments to us at biosci-help [At] net.bio.net