Beruflich Dokumente
Kultur Dokumente
CMBR 200?
4NBSU4FOTPST
Compact cameras for
machine vision
"VUPNPUJWF
"TTFNCMZ
Pobotic workceIIs in
auto-racking appIications
#JPNFUSJDT
NoveI systems tackIe
fingerprint anaIysis
#VTJOFTT7JFXT
John NagIe on working
with system integrators
7*4*0/"/%"650."5*0/40-65*0/4
'03&/(*/&&34"/%*/5&(3"5034803-%8*%&
7JTJPO
4ZTUFNT
% & 4 * ( /
7JTJPO
4ZTUFNT
% & 4 * ( /
www. vi si on- syst ems. com
-BCFM
*OTQFDUJPO
Vision system performs
DCP verification
HALCON 8.0 +++ One number +++ Many possibilities
Apply intelligence.
And discover speed.
In the food industry, speed and precision matter most.
HALCON 8.0 has everything under control from fill level
by label position to bar code legibility. Accurate as it is,
HALCON 8.0 also loves to go fast: up to 100,000
not miles, but bottles per hour. And of course it also
handles other food and packaging. HALCON brings
more intelligence to your industry.
For more information: www.halconists.com
A product of MVTec
0712VSD_C2 C2 11/29/07 11:22:56 AM
2007 National Instruments Corporation. All rights reserved. National Instruments, NI, and ni.com are trademarks of National Instruments.
Other product and company names listed are trademarks or trade names of their respective companies. 2007-8608-221-101
800 433 3488
>>
Get your FREE 30-day trial version at ni.com/vision/vbai
Vision Builder AI NEW!
Configurable
Machine Vision
No Programming
Required
Acquisition
from Thousands
of Cameras
Inspect Faster
with NI Vision Builder AI
Make your vision a reality. National Instruments Vision Builder for Automated
Inspection (AI) now features an innovative state machine editor, taking you from
initial design to deployed vision application faster than ever. Take advantage of
the more than 100 included image processing steps, acquisition from thousands
of cameras, and easy integration with existing industrial control hardware to make
your deployment faster and easier no programming required.
0712VSD_1 1 11/29/07 10:48:39 AM
Headquaters: The Imaging Source Europe GmbH
Sommerstrasse 36, D-28215 Bremen, Germany
Phone: +49 421 33591-0
North & South America: The Imaging Source, LLC
7257 Pineville-Matthews Road, Charlotte, NC 28226, USA
Phone: +1 704-370-0110
Asia Pacific: The Imaging Source Taiwan Co. Ltd.
6F C-3, No. 58, Jhouzih Street, Taipei City 114, Taiwan
Phone: +886 2-2797-8508
www.theimagingsource.com
Software included
Driver
for WDM, DirectX
and MIL
under Windows
and
a C++ Class Library running under Windows
2000,
XP and Vista
FireWire cameras
Sony progressive scan CCDs
Trigger and I/Os
640 x 480,1024 x 786, 1280 x 960 pixel
Up to 60 image/s
C- and CS-mount
IEEE 1394a
USB cameras
Sony progressive scan CCDs
Trigger and I/Os
640 x 480,1024 x 786, 1280 x 960 pixel
Up to 60 image/s
C- and CS-mount
USB 2.0
USB + FireWire Cameras
0712VSD_2 2 11/29/07 10:49:07 AM
VisionSystems
D E S I G N
D e c e mb e r 2 0 0 7
A PENNWELL PUBLICATION
VOL. 12 NO. 12
Vision and Automation Solutions for
Engineers and Integrators Worldwide
Feat ur es
On t he Web www.vision-systems.com
www. v i s i o n - s y s t e ms . c o m Vi s i o n S y s t e ms De s i g n De c e mb e r 2 0 0 7 3
Depar t ment s Col umns
Business Views
Interview with
John Nagle,
Nagle Research
Pg. 9
13 Technology Trends
IMAGE PROCESSING System
speeds package production
BIOMETRICS Vision systems
tackle ngerprint analysis
LABEL TRACKING RFID and
vision team for pharmaceutical
packaging
QUALITY CONTROL Machine
vision checks bottle-cap seals
QUALITY CONTROL Vendors
benet from document-validation
systems
6 Letter to the Editor
9 Business Views
47 Vision+Automation
Products
55 Ad Index/Sales Ofces
Complete Archives
Industry News
Buyers Guide
Webcasts
White Papers
Feedback Forum
Free e-newsletter
Video Library
25
Industrial Automation Products:
Developing vision-guided robotic work-
cells Robot, smart cameras, lighting,
and PC teamed in auto-racking
application. Valerie Bolhouse
35
Prole in Industry Solutions:
OCV supports FDA-compliant
packaging line Multiple machine-
vision stations inspect and verify
pharmaceutical-vial packaging
system. Winn Hardin
39
Product Focus: Sensors tackle
machine-vision applications Vision
sensors are closing the gap between
traditional photoelectric sensors and
more complex image-processing
systems. Andrew Wilson
5 Inside Vision
Signed, sealed,
delivered
56 My View
Blue rinse group
Cover St or y
Packaging device integrates
packaging, barcoding, and
labeling systems. (See p. 13;
photo courtesy SuperUser
Solutions)
See pg. 39
0712VSD_3 3 11/29/07 10:49:29 AM
The IVC-3D Smart Camera brings the advantage of seeing all three dimensions for location
and inspection of any part. True three-dimensional shape is the key for optimal picking
precision and opens a new range of efcient robot solutions on the factory oor. A new
dimension in vision-based robot guidance has emerged!
For more information, visit www.sickivp.com or www.sick.com
SICK IVP is a part of the SICK Group providing industrial vision cameras for factory automation.
: Innovation
IVC-3D A new dimension in Robot Vision!
SICK AG | Waldkirch | Germany | www.sick.com
0712VSD_4 4 11/29/07 10:52:02 AM
InsideVision
5 www. v i s i o n - s y s t e ms . c o m Vi s i o n S y s t e ms De s i g n De c e mb e r 2 0 0 7
Signed, sealed, delivered
T
rade shows can be dangerous places. Tey can overwhelm a visitor with sensory
input as vendors compete for attention with ashy booths, pulsing music, and at-
tractions such as food and drink. Shows in the machine-vision industry have been
notable for their lack of these marketing lures, but times are changing. VISION
2007, held in Stuttgart this past November, certainly opened new realms for the industry,
with its expanded oor space that now encompasses two vast halls of the New Stuttgart
Trade Fair Centre. Te show drew 280 exhibitors, up 30% from last year, and more than
6000 visitors, a 13% increase. Te show had lots of ash and, fortunately, most of the sub-
stance that a major show should have, delivering many innovative new products. To show
how new technologies and products could be used in machine-vision applications, many
companies displayed systems using these products. Our coverage of the technical develop-
ments, new products, and applications at the show will appear in our January issue.
Meanwhile, the topic of packaging dominates our coverage in this issue. Pack Expo
held in Las Vegas, NV, in Octoberdrew 1200 exhibitors and 25,000 visitors. Although
the dierence in scale was stark between the two trade shows, the importance of machine
vision to the packaging industry is apparent.
VALIDATING VISION
Our cover story by editor Andy Wilson describing an automatic labeling and verica-
tion system from SuperUser Solutions is the rst in a series of articles on machine vision
in packaging. Subsequent articles on pharmaceutical packaging show how RFID and
vision can be teamed to track drugs and inspect the documents that accompany many
drug packages. A feature article by contributing editor Winn Hardin describes a multi-
camera and barcode-reader inspection line from Systech International that helps manu-
facturers meet US FDA requirements. For another perspective, visit our Web site to view
a recent webcast by David Dechow of Aptra Machine Vision Solutions describing how
machine-vision systems are being used for container and packaging inspection.
One of the tools that has been driving the acceptance of machine vision in packaging
systems is the vision sensora relatively low-cost device that uses embedded software. Our
Product Focus by editor Andy Wilson reviews the growing number of these products and
how they are being integrated into a new generation of production equipment.
How to integrate machine vision into a more sophisticated application such as a robot
workcell is the topic of an article by Valerie Bolhouse, formerly at Ford Motor Company.
She relates the details of how an auto-racking system was developed for an automotive-
parts stamping plant. And, nally, system-integrator John Nagle, in our Business Views
interview, provides insight into how an end user or OEM should evaluate the prospect
of working with system integrators to develop complex 2- or 3-D vision systems.
Trade shows may be the place where these products and systems are put on display, but the
factory oor, minus the fanfare, remains one place where machine vision proves its value.
Kathy Bush: Publisher
Tel.: (603) 891-9434; e-mail: kathyb@pennwell.com
W. Conard Holton: Editor in Chief
Tel.: (603) 891-9161; e-mail: cholton@pennwell.com
Andrew Wilson: Editor
Tel.: (603) 891-9115; e-mail: andyw@pennwell.com
Bonnie Heines: Managing Editor
Tel.: (603) 891-9143; e-mail: bonnief@pennwell.com
Winn Hardin: Contributing Editor
e-mail: winn@hardingroup.com
David Lieberman: Contributing Editor
e-mail: davidlieberman@bellsouth.net
Adrienne Adler: Director of Marketing
Suzanne Heiser: Art Director
Dan Rodd: Senior Illustrator
Mari Rodriguez: Production Director
James Kirkland: Ad Services Manager
Debbie Bouley: Circulation Manager
EDITORIAL ADVISORY BOARD
Ben Dawson,DALSA IPD; David Dechow,
Aptura; Christian Demant, NeuroCheck
GmbH, Stuttgart, Germany; Prof. Dr. B.
Hfinger, Institute for Microelectronics
Stuttgart, Stuttgart, Germany; Toshi Hori,
GEViCAM; Henrik Ilsby, Copenhagen,
Denmark; Joseph A. Sgro, Alacron;
William Silver, Cognex Corp.; Wilhelm
Stemmer; Stemmer Imaging GmbH,
Puchheim, Germany; Nello Zuech,
Vision Systems International
EDITORIAL OFFICES: Vision Systems Design,
98 Spit Brook Road, Nashua, NH 03062-5737;
Tel: (603) 891-0123; FAX: (603) 891-9297;
http://www.vision-systems.com
CORPORATE OFFICERS:
Frank T. Lauinger: Chairman
Robert F. Biolchini: President and CEO
Mark Wilmoth: Chief Financial Ofcer
ADVANCED TECHNOLOGY DIVISION:
Gloria Adams: VP, Audience Development
Christine A. Shaw: Senior Vice President/
Group Publishing Director, Communications and
Optoelectronics Group
ATD PUBLISHING SERVICES:
Meg Fuschetti: Art Director
Judy Simers: Buyers Guide Director
Subscription inquiries: Tel.: (847) 559-7520;
Fax: (847) 291-4816; e-mail: vsd@omeda.com.
In Europe: Mailfast, JFK/BOS/850858, P.O. Box 66,
Hounslow, United Kingdom TW5 9RT;
Fax: 44 20 7504 8207
We make portions of our subscriber list available to
carefully screened companies that offer products and
services that may be important for your work. If you do not
want to receive those offers and/or information, please let
us know by contacting us at List Services, Vision Systems
Design, 98 Spit Brook Road, Nashua, NH 03062.
W. Conard Holton
Editor in Chief
cholton@pennwell.com
0712VSD_5 5 11/29/07 10:52:30 AM
Where do you want to be?
vhen it oones to naohine vision, there`s no need to work overtine.
Lse the probably easiest naohine vision software around - nade
by lV1eo. Oet innediate results with just a few nouse olioks. lt`s
easy, fast and eoononioal. Lnjoy your leisure tine - rely on lV1eo.
lor nore infornation see www.mvtec.com A product of
100
200
300
400
500
600
700
800
900
100 200
after smoothness, lambda: 0.010000
300 400 500 600 700
Letters
6 De c e mb e r 2 0 0 7 Vi s i o n S y s t e ms De s i g n w w w. v i s i o n - s y s t e ms . c o m
We Want to Hear From You
Send your letters to Conard Holton,
Editor in Chief, Vision Systems Design,
98 Spit Brook Road,Nashua, NH 03062 USA;
cholton@pennwell.com
David Storks article in the October issue
of Vision Systems Design (p. 69) is notable
for failing to mention that the artist Da-
vid Hockney and I have published a half-
dozen papers and one book on this subject
since 2000. Our work is widely acknowl-
edged as having established the power
of an artists visual skills for making
discoveries in art history, and the com-
puter techniques we developed could
signicantly aid the analysis of imag-
es in paintings. As a result, this has
now developed into a eld in its own
right, with the scientists who Stork did
mention subsequently applying their
own computer techniques to analyze
aspects of paintings.
Unfortunately, some of the work
in this eld has been misguided. As
has been known for more than a half-
century, results from a computer are
no better than the input data and
underlying assumptions. An article
written at the request of the editor
(IEEE MultiMedia 14 (2), 8 (2007))
details the false conclusions in three pa-
pers written by David Stork, all due to
errors in his data and assumptions, and
references in that article detail the awed
conclusions in a number of his other pa-
pers on this topic.
Since the mind and hand of the artist are
intimately involved in the creation process,
even when portions of a particular painting
are based on optical projections, these im-
ages are much more complex to analyze than
are photographs. However, as our results of
the past eight years show, a fundamental-
ly new approach to image analysis can be
developed when an optical scientist works
closely with a highly skilled artist.
Charles M. Falco
Chair of Condensed Matter Physics
and Professor of Optical Sciences
University of Arizona
falco@u.arizona.edu
Power of vision in art
C
o
m
p
u
t
e
r
a
n
a
l
y
s
i
s
b
y
K
i
m
o
J
o
h
n
s
o
n
a
n
d
D
a
v
i
d
G
.
S
t
o
r
k
;
2
0
0
7
M
u
s
e
u
m
A
s
s
o
c
i
a
t
e
s
/
L
A
C
M
A
.
0712VSD_6 6 11/29/07 10:42:57 AM
www.edmundoptics.com/VS
800.363.1992
Request your
FREE copy today!
BUILD THE
RIGHT SYSTEM
EVERY TIME
NEW LENS SELECTION GUIDE
www.edmundoptics.com/selectionguide
Match Imaging
Needs with Lenses
and Cameras
LARGE FORMAT LENSES
DOUBLE GAUSS
LENSES
SILVER SERIES
TELECENTRIC LENSES
HARSH ENVIRONMENT
OPTICS
w w w. v i s i o n - s y s t e m s . c o m Vi s i o n S y s t e ms De s i g n De c e mb e r 2 0 0 7 7
Tin-lm physicist Charles Falco nds it
notable that I did not cite his work (none
of which has passed rigorous peer review)
but my piece
1
was to promote the rst
International Symposium on Computer
Image Analysis in Art,
2
so I focused on
research by scholars who will present there
(see Vision Systems Design, Oct. 2007, p.
69). If I had been writing a historical over-
view, I would have cited many research-
ers who possess a scientists rigor and an
artists vision, such as Richard Taylor
a professor of physics and painter with
a masters degree in art, who pioneered
fractal image analysis of Jackson Pollocks
paintings a half-decade before Hockney
published his speculations. Surprisingly,
Falco feels his work with Hockney es-
tablished the power of an artists visual
skills for making discoveries in art histo-
ry, but everyone who has studied art his-
tory knows this rich tradition goes back
to artist Giorgio Vasaris Lives of the Art-
ists (1568) and earlier.
I am unaware, too, who widely ac-
knowledges the Hockney/Falco discover-
ies, given the fact that the unanimous con-
sensus in a four-day symposium and every
appropriate scholarly publicationby one
curator, seven historians of art and optics,
seven scientists and countingrejects their
highly promoted claim that artists as early
as 1420 secretly traced optical projections.
Moreover, rigorously peer-reviewed papers
in the relevant disciplines (computer vision
and pattern recognition) have pointed out
technical aws in Hockneys and Falcos
unpeer-reviewed methods.
3
Falco touts his collaboration with Hock-
ney, but their methodology is fraught with
problems. If two artists have dierent vi-
sions that t the evidence in a painting
equally well (as has happened numer-
ous times), which one is right? Science
can never appeal to fame or authority, of
course. Unless Falco gures out an objec-
tive way to prove to scientists that Hock-
neys beliefs are correct in such cases, then
Falcos resultseven if expressed in rig-
orous mathwill be mere interpretation
or speculation, not science.
Upset about the conclusion from sev-
eral scholars rebutting his theory, Falco
has complained to editors of at least 10
journals and conferences and the host in-
stitutions of at least three such scholars.
Falco suggests here and many places that
in publishing his rebuttal the editors of
IEEE MultiMedia had found errors in the
papers of at least three scholars, but he is
wrong. Sethuraman Panchanathan, its ed-
itor in chief, sets the record straight: It is
not accurate to suggest IEEE MultiMedia
was doing anything more than facilitating
a healthy exchange of ideas. Indeed, the
editors invited me and a coauthor to pub-
lish a rebuttal to Falcos claims.
4
To the
best of my knowledge, not a single such
editor or scholar has agreed with any of
Falcos protestations; moreover, every ex-
pert who has reviewed Falcos claims and
these scholars counter-arguments nds
his protestations without foundation.
Open debate, peer review, expert con-
sensus, and the rejection of statements that
appeal to authority have beenand will
always bethe proper scientic method
for determining truth.
David G. Stork
Chief Scientist, Ricoh Innovations
Visiting Scholar, Stanford University
artanalyst@gmail.com
1. D. G. Stork, Imaging technology en-
hances the study of art, Vision Systems
Design 12(10):69 (2007).
2. D. G. Stork and J. Coddington, eds.,
Computer image analysis in the study
of art, SPIE Press (2008).
3. For example, A. Criminisi and D. G.
Stork, Did the great masters use opti-
cal projections while painting? Perspec-
tive comparison of paintings and pho-
tographs of Renaissance chandeliers, J.
Kittler, M. Petrou, and M. S. Nixon,
eds., Proc. 17th Intl Conference on Pat-
tern Recognition IV, 645 (2004).
4. D. G. Stork and M. Duarte, Revisit-
ing computer image analysis and art,
IEEE MultiMedia 14(3):108 (cf., www.
diatrope.com/stork/FAQs.html; July-
September 2007).
Computer image analysis in the study of art
Te author replies
0712VSD_7 7 11/29/07 10:43:21 AM
F I R E F LY MV
F L E A 2
DRAGONFLY 2
OFTEN I MI TATED. NEVER DUPLI CATED.
The Dragonfly family has been serving OEMs reliably
since 2001, with thousands of cameras built into
imaging applications around the world.
HERE WHEN YOU NEED US
Easy access to responsive experts and
online resources to resolve technical
questions and issues quickly.
Works with industry standard C
and CS-mount lenses. Optional
microlens for space constrained
applications.
0. 3M, 0. 8M OR 1. 3M
Six different combinations of
monochrome and color 1/3
S C OR P I ON
TM
GR A S S HOP P E R
0712VSD_8 8 11/29/07 10:50:16 AM
w w w. v i s i o n - s y s t e m s . c o m Vi s i o n S y s t e ms De s i g n De c e mb e r 2 0 0 7 9
Business Views
VSD: What sort of systems or services
does Nagle Research provide?
Nagle: We have decided to devote ourselves
almost exclusively to 3-D machine-vision
development, allowing us to build an ex-
pertise and body of experience with 3-D
technology that is second to none in the
industry. Nagle Research is a SICK (Min-
neapolis, MN, USA; www.sick.com) vision
integrator. We are entirely brand loyal to
SICK|IVP vision products, most often us-
ing the Ranger series of cameras.
My business partner Andy Tyssen, also a
software engineer, and I started the compa-
ny in June 2003. Our rst project and what
is generally regarded as our claim to fame
is the Aurora automated high-speed railroad
track-inspection system. We have each spent
more than a decade making video games for
Nintendo, Playstation, and others. Tat ex-
perience has been immeasurably valuable in
keeping the performance of our
systems on the leading edge.
VSD: What should be asked when consider-
ing the services of a system integrator?
Nagle: It is impossible to engineer a solu-
tion without a thorough understanding of
the problem. But to truly know the prob-
lem, you have to get past the supercial
goals and get to the meat of the challenges
that a solution will have to face. Tere can
be many gremlins hiding below the surface
of what seems like an easy project.
For example, a candy factory needs a vision
system that can count jellybeans moving down
a conveyor belt. Tats the supercial goal. Ob-
viously this is a very straightforward task for a
vision system to accomplish. To be able to intel-
ligently plan a solution, however, requires much
more information. What should the system
do with the count? Does it need to trigger a
signal when a certain count is reached? Does
it need to communicate with a PLC? What if
a jellybean is malformed, does it count? And
how does the system deter-
mine what is a good jelly-
bean? How fast are the jelly-
beans moving? Do we need
to count the individual colors?
What are the space consider-
ations for the vision system?
Tis is very goal-orient-
ed fact-nding research,
and so this sort of question-
and-answer probing can be
done even by nontechnical
people. Once all of the major and minor
goals are known, then it is straightforward
to isolate the specic disciplines and skill
sets required to make the project a suc-
cess.
VSD: So what can be done in-house
by a company?
Nagle: Evaluating ones own capabilities or
the capabilities of company sta members
is the next step in deciding how much, if
any, of the project can be done in-house. If
the project can be accomplished with o-
the-shelf vision solutions or relatively sim-
ple smart cameras and only minor external
connectivity is required, then the chances
of being able to do this are good. If compli-
cated record keeping, PLC connectivity, or
advanced image-processing algorithms are
required, it is almost certain that a third-par-
ty vision-system integrator with software-de-
velopment capability will be necessary.
Dierent skills are required to integrate vi-
sion systems of varying degrees of complex-
ity. Even a good list of necessary skills can-
not be comprehensive and should be treated
only as a guideline or rule of thumb (see
table on p. 10).
VSD: What are the implications of
working with a 2-D vs. 3-D system?
Nagle: Most people who have experience
with vision are likely to have worked only
with 2-D systems. Two-dimensional sys-
tems deal with color and contrast; three-
dimensional systems deal with materials
and geometry. Te share a lot of the same
concepts, but, in general, 3-D is more dif-
cult to implement. Tis is because now
we are not just dealing with a light and a
camera, we have to deal with laser light fre-
quency; beam spread angle and thickness;
laser power requirements based on material
properties and stand-o distance; ranging
algorithms; angular orientation of cam-
era/subject/laser to obtain required accu-
racy; safety issues related to working with
the laser; and coping with less than ideal
material properties.
Integrating a SICK IVC-3D or a Ruler
product can mitigate some of these issues,
in that the camera lens, laser type, and ori-
entation are xed at the factory (which also
limits to some degree their applicability.)
Ranging algorithms and material proper-
ties must still be dealt with in any case.
Working with a machine-vision-system integrator
A discussion with John Nagle,
Nagle Research
JOHN NAGLE is president
and CEO of Nagle Research,
Cedar Park, TX, USA;
www.nagleresearch.com.
He is a professional software
engineer with more than 20
years experience developing real-time,
high-performance systems, including
more than four years of high-speed
3-D machine-vision development.
Editor in chief Conard Holton spoke
to him about how to evaluate the
need to hire a system integrator to
implement a machine-vision solution.
0712VSD_9 9 11/29/07 10:50:49 AM
Intelligent Cameras Area Scan Cameras Line Scan Cameras
click. see. smile! www.basler-vc.com/scout
30 Digital Camera Models with
GigE Vision and FireWire-b
Up to 100 meter cable lengths for more exibility
Enhanced bandwidth that allows even more cameras on one bus
Price/performance ratio at its best! Starting at 499 EUR
for a 60 fps VGA camera
Progressive scan CCD sensors ranging in size from
VGA to 2 megapixels
Easy to integrate with a footprint of 29 x 44 mm
Flexible and cost effective
De c e mb e r 2 0 0 7 Vi s i o n S y s t e ms De s i g n www. v i s i o n - s y s t e ms . c o m 10
BusinessViews
VSD: Is a vision software-development
kit di cult to learn?
Nagle: In any nonsmart camera system, the
integrator must have a thorough knowledge
of the vision hardware software develop-
ment kit (SDK), including the SDK for
the frame grabber, if applicable. Tese are
highly nontrivial software toolkits and a
deep-rooted foundation in C++ and soft-
ware development is essential. Even with
the requisite C++ experience, the SDK
itselflike any complex systemhas a
learning curve.
VSD: What are the benets of third-party
integration?
Nagle: Any competent vision integrator
should be able to integrate vision in sim-
ple to moderately complex projects. Many
vision integrators do not have great depth in
software and electrical engineering, and so
the more complicated vision projects are be-
yond them. When choosing an integrator,
it becomes important to match the skills
they bring to the table with the skills that
will be required. Dealing with an integra-
tor can save an enormous amount of time
and development eort. In many cases, ex-
perienced integrators have saved companies
from spending hundreds of thousands of
dollars on inappropriate equipment and
software.
For example, we were asked by a railroad-
equipment manufacturer to provide consul-
VI SI ON SKI LL SET / COMPLEXI TY RELATI ONSHI P
Basic understanding of lenses
Smart-camera systems
with only basic external
connectivity
Basic understanding of lighting for vision
Ability to owchart and use a PC
Working knowledge of Visual Basic
Simple 2-D/3-D systems
with limited external
connectivity
Professional-grade VB prociency (integrating
ActiveX, writing algorithms, and so forth)
Pro-level Prociency with C++ (clean, efcient
code, performance-tuned)
More complex 2-D/3-D
systemshigher speeds,
more connectivity
multiple cameras, and
high-speed image
processing
Pro-level understanding of the vision hardware
(operating modes, speeds, and so forth)
Pro-level understanding of communication
protocols (UDP/TCP, RS232, and so forth)
Ability to generate logic-level signals from PC,
either via ports or electrical circuit
Pro-level understanding of optics (lenses, lters,
lights, lasers, and so forth)
0712VSD_10 10 11/29/07 10:51:14 AM
> COMMON VISION BLOX The freedom to develop your imaging application your way.
www.commonvisionblox.com
> COMMON VISION BLOX enables you to concentrate on what matters most: creating your
application! CVB offers you the functionality, performance and reliability you expect from
an industrial product. Enjoy shorter development times and a robust solution delivering
100% accuracy. Field proven, with more than 20.000 licenses deployed worldwide and from a
company with over 30 years experience. And of course, a worldwide partner network always
on-hand to offer local support and service. Imaging is our passion.
picture a programming library
as broad as your vision.
w w w. v i s i o n - s y s t e m s . c o m Vi s i o n S y s t e ms De s i g n De c e mb e r 2 0 0 7 11
BusinessViews
tation as to what camera would be required
for a 2-D high-speed-railroad-inspection
system. Te company had already spent
many thousands of dollars on image-pro-
cessing software to locate defects in cross-
ties using 2-D imaging techniques. Te
problem was that their approach had not
accounted for surface stains, sealant, and
debris confusing the analysis software. We
ultimately concluded that a 3-D solution
was more appropriate for this application
and developed a Ranger-based solution that
handles these material properties nicely.
VSD: When working with a system
integrator, what are you paying for?
Nagle: Speaking only for Nagle Research,
in most cases vision projects are quoted
on a at fee basis. Usually the process is
phone conference to discuss the challenges
and goals; if possible, samples are sent for
testing and proof-of-concept, If the project
proves solvable, we submit a proposal.
With projects whose goals are a moving tar-
getfor example, additional defects to detect
or additional accuracy requiring more cam-
erasthere will most likely be proposed a at
fee for a dened scope of work and a standard
hourly fee for work that is out of scope..
For our fee, the client receives our profes-
sional consultation, software and electrical
engineering resources, and a solution that
meets their requirements. In most cases, un-
less specically agreed to, the client does not
get source code to the nal solution. In some
arrangements we will relinquish source code
for the application, for example, their user in-
terface and project-specic algorithms. Our
proprietary Javelin Vision Engine, however,
remains closed source. Javelin is the 3-D tech-
nology infrastructure to help us in developing
more robust vision systems
VSD: What are the fundamental questions
to ask before calling an integrator?
Nagle: Te basic questions that need to
be answered before an integrator is called
are
Is the project outside the scope of
in-house capabilities?
Is the company open to using third-
party integrators?
What is the price of failure or delays
arising from lack of internal experience?
Is there a budget for vision that
includes third-party integration?
Is there likelihood that given a
workable solution within budget,
the project would proceed?
If the answer to all of these is yes, then
most any integrator would be willing to
take the challenge. A competent integra-
tor is key to a sucessful system.
Whether or not that expertise comes
from within or from a third party is a deci-
sion the client ultimately will have to make.
Te most important thing is that a broad
skill set and expertise in a variety of disci-
plines will be required to complete the proj-
ect on time and on budget.
0712VSD_11 11 11/29/07 10:51:36 AM
Main Entry:
1
FireSync
I MAGE PROCESSI NG
SuperUser Solutions automatic top labeling system can print,
label, and verify packaged labels. Using an embedded PC
smart HMI user interface to control the printing mechanism,
applicator, and barcode verication system, the labeling
system can achieve speeds of up to 45 packages per minute.
0712VSD_13 13 11/29/07 10:43:54 AM
De c e mb e r 2 0 0 7 Vi s i o n S y s t e ms De s i g n w w w. v i s i o n - s y s t e ms . c o m 14
TechnologyTrends
Conventional ngerprint systems process
inked ngerprint cards that have been
manually scanned or ngerprints that
are electronically digitized with atbed
scanners. To date, hundreds of millions
of such prints have been collected and en-
rolled into systems such as the US Federal
Bureau of Investigation (FBI) Integrated
Automated Fingerprint Identication Sys-
tem and the US Department of Homeland
Security (Washington, DC, USA; www.
dhs.gov) Automated Biometric Identica-
tion System.
To digitize all ten prints from the dig-
its of the hands, todays capture tech-
nology requires approximately three
minutes. In 2004, the Department of
Homeland Security instructed the FBI
to expand its ngerprint database, man-
dating the use of faster scanning times.
Tat same year, the National Institute
of Justice (NIJ; Washington, DC, USA;
www.ojp.usdoj.gov/nij) solicited applica-
tions for a fast ngerprint-capture pro-
gram to develop more e cient methods
for collecting ngerprints.
Te NIJ has two basic requirements
for fast ngerprint-capture systems, says
Laurence Hassebrook, associate profes-
sor of electrical and computer engineer-
ing at the University of Kentucky (Lex-
ington, KY, USA; www.uky.edu). First,
the system must scan all the digits of one
hand in less than 10 s, and second, it must
generate an image of a rolled-equivalent
scannger nail to nger nailat 500
pixels/in. or better without the help of a
human operator.
Four independent projects for live-scan
replacement are included under the NIJ
Fast Fingerprint Capture Program us-
ing sensor technologies that are consid-
erably dierent from either ink or atbed-
scanner based systems. At Cross Match
Technologies (West Palm Beach, FL,
USA; www.crossmatch.com), the com-
pany is developing a U-shaped exible
polymer-foil-based substrate with a sen-
sor that conforms to the shape of each
nger. Using a dierent approach, TBS
North America (Herndon, VA, USA;
www.tbsinc.com) is developing a circu-
lar optical mirror system that the ngers
are drawn across, creating an image. At
the International Association for Identi-
cation Exhibition and Conference (July
2428; San Diego, CA, USA) the com-
pany announced a single-nger version of
the product, known as the Touchless Ten
Printer, capable of digitizing 10 rolled
equivalents in less than 20 s. A prototype
device capable of multiple nger digitiza-
tion is expected soon.
Finally, both Carnegie Mellon Universi-
ty (CMU; Pittsburgh, PA, USA; www.cmu.
edu) and the University of Kentucky are de-
veloping camera-based systems that capture
the 3-D shape of the hand and the friction
BI OMETRI CS
Vision systems tackle ngerprint analysis
and at runtime transferred to the host G3
HMI over USB, Ethernet, or manually
using ash memory. For simple applica-
tions, says Messineo, the software allows
you to add, edit, and delete modules such
as PLCs, motor drivers, photoelectric sen-
sors, and barcode readers and map data col-
lected from them to other external devices
on the network.
To congure the PresencePLUS sensor
for this application, it is interfaced to a PC
running PresencePLUS PC software, says
Messineo. Ten system settings, the bar-
code inspection, and the communication
interface must be set before the sensor is in-
terfaced to the Red Lion HMI. After con-
guring the sensor to operate over Ethernet,
the Banner vision sensor can automatically
transfer discrete I/O and inspection pass/
fail information over Ethernet. In this case,
however, it was necessary to transfer vision
tool results to the Red Lion HMI. Because
of this, a communication tool, available as
part of the PresencePLUS software, was
used to export measured barcode infor-
mation to the HMI over an Ethernet in-
terface, Messineo says. It allows you to
adjust printer and applicator parameters re-
motely over any Internet connection.
Four independent projects for live-scan
replacement are included under the NIJ
Fast Fingerprint Capture Program using
sensor techno logies that are consider-
ably different from either ink- or atbed-
scanner-based systems. These include
a exible foil contact sensor from Cross
Match Technologies (top left), a Hand-
Shot identication system from CMU
(top right), and prototype nger and
hand readers from the University of
Kentucky (bottom right).
0712VSD_14 14 11/29/07 10:44:16 AM
0712VSD_15 15 11/29/07 10:44:37 AM
The Experience of more than 5,000
Applications World-wide.
NeuroCheck GmbH : Software Design & Training Center
70174 Stuttgart / Germany : Phone +49 711 229 646-30
NeuroCheck GmbH : Engineering Center
71686 Remseck / Germany : Phone +49 7146 8956-0
e-mail: info@neurocheck.com
NeuroCheck is the efcient solution for all application
areas of image processing in manufacturing and quality
control. More than 1,000 library functions, congured
by mouse click in every conceivable combination, help
you to rapidly create cost effective and reliable solutions
for the entire eld of industrial visual inspections. Your
benet: shorter realization time, company-wide standard-
ization, and greater reliability compared to conventional
programming. NeuroCheck represents a consistently
integrated concept from the software through to the
complete application including all vision components.
PLUG & WORK!
For more information: www.neurocheck.com
TechnologyTrends
w w w. v i s i o n - s y s t e ms . c o m
ridges of the ngers and palms. CMUs sys-
tem, the Hand Shot ID system, constructs
a 3-D model of both hands by stitching im-
ages from multiple cameras.
To simultaneously obtain 1000-pix-
el/in. images of all 10 rolled-equivalent
nger and both palm ridge patterns and
minutiae within less than 5 s, the Hand-
Shot ID System uses multiple cameras
and spotlights to eliminate the need for
glass platen imaging. For HandShot to
accurately and instantly capture and re-
cord friction ridge skin detail on 10 rolled-
equivalent ngerprints and both palm
prints within 5 s, says Latanya Sweeney
of CMUs Data Privacy Lab, the system
constructs a 3-D model of both hands,
including palms, ngerprints, ngertips,
and sides of the ngers. Image-processing
algorithms then stitch images from mul-
tiple cameras together to form a complete
3-D model of both hands, extract ridge
detail, and translate the 3-D images to
standard formats.
Rather than use spotlights to illumi-
nate the hand, Hassebrook and his col-
leagues at the University of Kentucky re-
ly on structured light. By projecting and
capturing a series of striped patterns over
an object, its shape can then be deduced
by analyzing the way the stripes warp over
the objects surface when viewed at an an-
gle by a camera. In Hassebrooks system,
the hand is scanned, a 3-D image gener-
ated in real-time and then converted to
simulate a 2-D rolled ngerprint.
Developed in conjunction with Flash-
Scan 3-D (Austin, TX, USA; www.
ashscan3D.com), the initial prototype
scanned a single nger. Te subject plac-
es a nger over an opening in the systems
enclosure. A digital light projector projects
a series of striped patters onto the ngers
ridges. An array of three Camera Link cam-
eras from Basler (Ahrensburg, Germany;
www.baslerweb.com) then acquires a se-
ries of images that wrap around the nger.
Tese images are transferred to a host PC
using two Solios XCL video-capture cards
from Matrox Imaging (Dorval, QC, Can-
ada; www.matrox.com/imaging).
A second prototype uses a single 4M-
pixel Camera Link camera with a single
Matrox Helios capture card to acquire
a scan region large enough to capture a
human subjects palm. Te system can
also acquire the prints of all four ngers
simultaneously but does not achieve
wraparound scanning of any nger. To
achieve instantaneous acquisition, future
prototypes will feature a single, contin-
uously projected custom designed com-
posite pattern instead of the projected
striped patters and higher-resolution
cameras for simultaneous acquisition of
the entire hand with wraparound nger
scanning. After the system is completed,
it will be tested by the Kentucky State
Police and then delivered to the NIJ for
further evaluation.
0712VSD_16 16 11/29/07 10:44:54 AM
hi-def from any point of view
For all the details call 1-888-880-8474 or visit www.panasonic.com/medical
Now you can shoot HD images in places you simply couldnt before.
With Panasonics 16x9 multi format 1080i / 720p selectable high
definition remote head camera. Its the true HD remote head camera
you've been looking for.
Three 1/3" 16x9 HD sensors with native progressive scan running at
60 fps combined with a newly developed DSP offering 14-bit A/D
conversion and 19-bit inner processing deliver stunning detailed
images. Precise color reproduction is achieved via 12-axis matrix
control with independent color adjustment. And a unique dynamic
range feature makes highly contrasted scenes maintain their color
and detail in both dark and bright areas. You simply wont find bigger
performance in such a small camera.
Its true HD technology only from Panasonic. Now from any point of view.
lens not included
0712VSD_17 17 11/29/07 10:45:30 AM
De c e mb e r 2 0 0 7 Vi s i o n S y s t e ms De s i g n w w w. v i s i o n - s y s t e ms . c o m 18
TechnologyTrends
Shipping the correct product in the proper
package is vitally important in the phar-
maceutical industry. Should drugs be
wrongly packaged, the patient could suf-
fer serious medical complications. Tus, it
is vital that each drug be properly pack-
aged and labeled. However, since similar
containers are often used by pharmaceuti-
cal vendors, proper labeling alone may not
guarantee that the correct drug is placed
in the package.
To overcome these limitations, suppli-
ers of such containers are now embedding
radio frequency identication (RFID)
ICs into the containers before shipping
them to pharmaceutical vendors. By pro-
gramming these RFID tags with the type
and strength of each drug and other rel-
evant data, pharmaceutical vendors are
providing their customers with a secure
way of packaging the correct product in
the proper package.
RFID tags provide only one level of secu-
rity, says Brian Mack, sales engineer with
CIVision (Aurora, IL, USA; www.civision.
com), a manufacturer of machine-vision
systems. As well as reading the RFID tags,
each package must be inspected to ensure
that the information contained in the bar-
code label on the package corresponds with
that of the RFID tag.
CIVision has developed a pharmaceuti-
cal-packaging-inspection system, known
as LOMAX RFID, using o-the-shelf ma-
chine-vision components (see gure). PC-
based, the system inspects pharmaceutical
packages as they move along a conveyor
belt and into a specially designed housing
that accommodates the lighting, RFID
reader, and CCD cameras.
As containers enter this housing, a
Cobalt HF reader from Escort Memory
Systems (EMS; Scotts Valley, CA, USA;
www.ems-rdj.com) reads the informa-
tion stored in the RFID tag. Because
tags emit signals with power levels on
the order of millwatts, the readers sen-
sitivity is paramount, says Mack. De-
velopers must also take into account
the receive noise, interference, and oth-
er eects that may increase a readers bit-
error rate the ratio of the number of
bits incorrectly received to the total num-
ber of bits sent in a specied period. Te
EMS Cobalt HF reader, with 77-dBm
sensitivity, provides enough power to acti-
vate a tag buried inside a pallet of stacked
cases, he says.
After the Cobalt HF reader checks each
tag, data from each tag are transferred to
the host PC over an Ethernet interface. To
compare this information with the two-
dimensional Data Matrix code located
on each container, the barcode of each la-
bel is then read. As containers move along
the conveyor, they are illuminated by a 4
0.5-in. FL201 area red LED front light
from Metaphase Technologies (Bensalem,
PA, USA; www.metaphase-tech.com), and
an image of the package is captured us-
ing an A640 640 480-pixel gray-scale
Gigabit Ethernet camera from Basler
Vision Technologies (Ahrensburg, Germa-
ny; www.baslerweb.com).
To transfer captured images to the host
PC, CIVision Engineers used a Solios
GigE interface card from Matrox Imaging
(Dorval, QC, Canada; www.matrox.com/
imaging) that o oads the GigE Vision
protocol and reconstructs images from
transmitted data packets and passes the
resulting image to the computer host. Ac-
cording to CIVisions Mack, this frees
up host resources and avoiding interrupt
loading on the PC. Just as with its suc-
cessful LOMAX NB (Neck and Bottle
Inspection) CIVision used the Matrox
Imaging Library (MIL) to capture and
process these images(see Vision Systems
Design, February 2007, p. 19).
Using MIL measurement functions, a re-
gion of interest around the barcode of each
package is isolated and then read. Tese
data are then compared with the data col-
lected from the RFID system and the re-
sults of the comparison displayed on the
operators console. Once again, the compa-
ny used its CIVCore, a user interface with
MIL low-level vision processing, to control
the inspection process.
As each part emerges from the inspec-
tion station, packages where both barcode
and RFID information can be rejected by
pneumatically controlled actuators using a
PLC programmed from the host PC. With
the ability to inspect approximately 400
packages per minute on a single production
line, the LOMAX RFID can be supplied
in a number of dierent congurations, ac-
cording to Mack. Pricing of each system
is dependent on a number of factors, in-
cluding the type of product and the speed
required, he adds.
LABEL TRACKI NG
RFID and vision team for pharmaceutical packaging
LOMAX RFID system combines off-the-
shelf machine-vision components to
inspect parts at more than 400 packages
per minute (left). Software from Matrox
reads Data Matrix codes and compares
the data from embedded RFID tags; the
results are displayed on a graphical user
interface (below).
0712VSD_18 18 11/29/07 10:45:56 AM
Make the switch to a new generation
Americas: +1 800 445-5444 (toll-free)
Europe, Middle East & Africa: +45 4457 8888
Asia Pacijc: +81 45-440-0154 www.jai.com/switch
One core concept. One interface.
One simple choice.
C3 Camera Suite Unlimited Digital Switchability
All cameras in the three-tiered C3 Camera Suite
use the same interfacing connectivity and
share a common design philosophy.
From the entry-level Compact to the versatile Basic
and the high-end, pre-processing-packed Advanced
tier, it is incredibly easy to switch from one to another
to meet specijc deployment needs.
Swi t ch
0712VSD_19 19 11/29/07 10:46:20 AM
Innovation at Work
Participate in the preeminent event focused on the future of imaging.
Get an insiders look at the technologies and industries that are having a
huge impact on humankind. Meet the leaders who can help you with
your research and your career. Find the right teaming opportunities to
take your projects to the next level.
Dont miss being part of the excitement at Electronic Imaging 2008.
Sponsored by
IS&T
L
e
f
t
p
h
o
t
o
:
C
o
u
r
t
e
s
y
o
f
P
a
c
i
f
i
c
N
o
r
t
h
w
e
s
t
N
a
t
i
o
n
a
l
L
a
b
o
r
a
t
o
r
y
.
M
i
d
d
l
e
p
h
o
t
o
:
b
y
S
t
a
n
l
e
y
L
e
a
r
y
.
Conference: 2731 January 2008
San Jose Marriott and
San Jose Convention Center
San Jose, California USA
IS&T/SPIE 20th Annual Symposium
electronicimaging.org
De c e mb e r 2 0 0 7 Vi s i o n S y s t e ms De s i g n w w w. v i s i o n - s y s t e ms . c o m 20
TechnologyTrends
To meet supplier demands, manufacturers
of bottled products must ll, cap, and seal
products rapidly using high-speed systems.
Ten each of these products must be checked
to determine whether the bottle has been
lled to an acceptable level and whether the
cap is correctly placed and positioned.
Today, says Steve Belling, product
manager with Silgan Equipment Com-
pany (Downers Grove, IL, USA; www.
silganequipmentcompany.com), these
manufacturers are also incorporating add-
ed levels of protection in the form of tam-
per bands that are wrapped around the cap.
With tamper bands in place at the base of
the closure, consumers are assured that the
product has not been altered. Furthermore,
such sealing also guarantees the quality and
freshness of the bottled product.
While such packaging techniques may
provide the consumer with an added level
of security, they present the system
integrator tasked with high-speed in-
spection an extra level of complex-
ity. Not only must the automated
inspection system check for correct
ll level and proper cap placement,
it must also inspect the integrity of
the tamper band. At this years Pack
Expo (Las Vegas, NV, USA), Silgan
Equipment presented its latest bot-
tle-inspection system capable of per-
forming all three tasks at speeds of
up to 1200 bottles/min (see gure).
Conveyor-based, the Silgan sys-
tem uses an embedded host PC and
a PLC to control all the functions of
the machine, including the conveyor
speed, lighting control, camera inter-
face, triggering mechanisms, and re-
jection mechanisms. In operation, bottles to
be inspected move along the conveyor where
they are illuminated by a 2.5-in.-square
white at-panel LED. After their presence
QUALI TY CONTROL
Machine vision checks bottle-cap seals
Capable of inspecting as many as 1200 bottles/min,
Silgans bottle-cap-inspection system uses multiple
cameras to provide 360
o
views of each bottle as it
passes through a vision-based inspection station.
0712VSD_20 20 11/29/07 10:46:46 AM
ARVOO Imaging Products BV | The Netherlands | Phone +31 348 413 897 | Fax +31 348 417 242 | www.arvoo.com
visit www.arvoo.com, mail to sales@arvoo.com or contact your local distributor
The Axioma
videoprocessor
(AX-02)
model
AX-02-VGA
model
AX-01
compact integration of
framegrabber and computer
powerfull parallel video
processing architecture
independent processing nodes,
network connected
strong FPGA/PowerPC
optimized combination
targeting demanding
applications
N
E
W
from the inventor of opticlink
QUALI TY CONTROL
Vendors benet from
document-validation systems
0712VSD_22 22 11/29/07 10:47:37 AM
High Operability
High Efficiency at 900~1700nm
Uncooled / TE cooled
1x512, 320x256 and 640x512 array size
Ceramic leadless chip /
Koavar metal case packaged
Near-infrared imaging
Imaging spectroscopy
Covert surveillance
Semiconductor inspection
Medical science and biology
Fiberoptic telecommunication
Astronomy and scientific
Industrial thermal imaging
)HDWXUHV
)HDWXUHV )HDWXUHV
$SSOLFDWLRQV
$SSOLFDWLRQV $SSOLFDWLRQV
,Q*D$V
,Q*D$V
)RFDO3ODQH$UUD\V
)RFDO3ODQH$UUD\V
Your best NIR/SWIR imaging and spectroscopy solutions Your best NIR/SWIR imaging and spectroscopy solutions
Email:cchong@cht.com.tw
http://leadinglight.chttl.com.tw
w w w. v i s i o n - s y s t e m s . c o m Vi s i o n S y s t e ms De s i g n De c e mb e r 2 0 0 7 23
TechnologyTrends
200% manual inspection rule, requiring
two individuals to compare the two docu-
ments. Recently, the FDA has standardized
the initial original New Drug Application
submission format to an XML structure,
but the conversion to the insert requires the
same proofreading intensity to improve ac-
curacy of the nal printed material. But, un-
like the standard drug facts on label copy,
the type size, paper, and print color and con-
trast may still make reading di cult. Now,
says Parish, with the introduction of the
FDA 21 CFR Part 11 requirements, regula-
tions are in place for electronic records and
signatures to also ensure pharmaceutical
manufacturers must validate that the docu-
ment and revisions are correctly identied as
well as containing the correct copy.
With documents so highly complex, proof-
readers may spend many hours on each docu-
ment to ensure that each one is correct. Add
in multiple languages such as Japanese, Greek,
or Chinese, and the task is almost impossible.
Even with such people in place, says Parish,
any human error such as a misplaced deci-
mal point can result in costly recalls for the
pharmaceutical manufacturer or vendors or
a fatal dosage for the patient.
CIS has developed a number of auto-
mated documentation systems and soft-
ware suites that can scan, recognize, and
automatically detect any errors between an
original electronically generated document
and a proof or press version. Te suite of
software, known as AutoProof Pro, consists
of a number of dierent modules.
Most original documents are created by
the manufacturer and sent to a printer. Te
printed inserts are shipped to the pharma-
ceutical manufacturer for incoming 200%
inspection. To reduce the inspection process
by up to 95%, CIS provides an image-com-
pare module integrated with several input
devices. For comparing materials up to 11
17 in., a atbed scanner is used. For larg-
er material, a sheet-fed scanner that can ac-
commodate up to a 54 54-in. documents
at up to 1200 dpi is used. After images are
scanned, Docu-Match software auto-aligns
the two images and nds any dierence in
text, images, and color. Te software auto-
matically aligns the master and incoming
material for accurate comparison.
In many cases, says Parish, the pre-
press version must be compared with a
document generated as an Adobe Acro-
bat PDF le. To do this, the PDF is con-
verted to a bitmap and compared at the
same resolution as the scanned materials.
For comparing electronic revisions, CIS
can compare each character, character for
character, even though the format and lo-
cation of the copy has moved. Results of
the dierences between digitized and orig-
inal documents can then be compared on
the PCs monitor (see gure).
Many of these vendors produce drugs
for clinical trials, for which small batches
of labels must also be checked for accuracy.
For these companies, CIS has developed a
programmable vision workstation that us-
es a digital camera with the same software
suite to check labels printed on laser, ther-
mal-transfer, or dot-matrix printers. To do
this, the camera captures each label to detect
defects as small as 0.005 in. and uses optical
character recognition to read multiple areas
on the labels and compare the data with that
from clinical-label generation program.
Says Parish, We generally oer a to-
tally integrated system including com-
puter, cameras, scanners, and software to
simplify validation and use. It has already
been adopted by several pharmaceutical,
prepress, and printing companies, which
need to rapidly and accurately check high-
ly complex documents.
By registering original pharmaceutical
documents with prepress versions,
pharmaceutical vendors are using soft-
ware from CIS to proofread text, bar-
codes, graphics, and molecular diagrams.
0712VSD_23 23 11/29/07 10:47:58 AM
w w w . e u r e s y s . c o m
Download
Develop
Deploy
Open eVision 1.0
Instantaneous web download or CD delivery
- Modern, fast and simple installation
Software-only licensing scheme
- Automatic on-line or of-line activation
- License Manager for licensing administration
High-value Software Development Kit
- Open eVision Studio
- Extensive documentation and sample programs
- Support of main or auxiliary development systems
Deployment facilities in production environments
IEEE
1394
GigE
Vision
USB
2.0
Frame
Grabber
0712VSD_24 24 11/29/07 11:01:34 AM
IndustrialAutomation Products
w w w. v i s i o n - s y s t e m s . c o m Vi s i o n S y s t e ms De s i g n De c e mb e r 2 0 0 7 25
A
hot application for machine vi-
sion right now is vision-guided
robotics for automatic loading
and unloading of shipping racks.
Te benets provided to manufacturers for
automating this application make the job
of cost justication easy. A manufacturing
line fully automated except for
the load/unload process will
be constrained by the opera-
tor tending the line. And since
many times the parts are heavy
and awkward, the safety of the
operator is often a concern.
A typical shipping rack in an
automotive stamping plant has
a footprint of 4 8 ft and will
hold about an hours worth of
production, depending on the
complexity of the subassembly
or the size of the parts. A fork-
lift driver will pick up a full
rack and drop o a new empty
rack in station in a loosely de-
ned location.
Shipping racks are not preci-
sion tooling. Tey are designed
to provide maximum protection to the prod-
uct at a minimum cost, since there might be
100 or more racks in the system at one time.
Tey are handled roughly, jostled during tran-
sit in the trailer, and can be stored outside. An
automated cell to load the racks has to take all
of these factors into consideration.
THE PROBLEM
Lets look at what goes into a completed
solution for an auto-racking application.
Well do this in the context of a case study
of a system successfully integrated and in-
stalled in a stamping plant. Te application
is loading completed side sill assemblies in-
to a rack. Te side sill is a sheet-metal
stamped part about 6 ft long weigh-
ing 25 lb. Te rack has seven hooks
on each side and each hook holds sev-
en parts. When completely lled, the
rack is rotated and the other side is
loaded (see Fig. 1).
Besides nding the overall location
of the rack, the vision system has to
ensure the rack is not damaged and ac-
ceptable to load without a crash.
If the shipping bar is up, the
part will crash into
the rack. If an arm
is bent down, the
robot will scrape
the part across
the bar and dam-
age the part .
On the surface
this type of appli-
cation appears to
be a fait accom-
plian automa-
tion solution that
can be purchased
o-the-shelf. But seeing
a demonstration of feasibility is
about 5% to 10% of the total engineer-
ing eort it takes to get a fully imple-
Developing vision-guided
robotic workcells
Robot, smart cameras, lighting, and PC teamed in auto-racking application.
By Val eri e Bol house
FIGURE 1. An auto-racking
application in a automotive
stamping plant required
that completed side sill
assemblies be loaded in a
shipping rack (top); a
drawing shows the rack
with two parts loaded and
the shipping bar (yellow)
in the up position (right).
VALERIE BOLHOUSE was
formerly a vision specialist at
Ford Motor Company, Detroit,
MI, USA. She has presented
ve Vision Systems Design
Webcasts on the Fundamentals
of Machine Vision, available
on demand at www.vision-
systems.com.
0712VSD_25 25 11/29/07 11:02:35 AM
De c e mb e r 2 0 0 7 Vi s i o n S y s t e ms De s i g n w w w. v i s i o n - s y s t e ms . c o m 26
IndustrialAutomation Products
mented system in your facility. So if you see
a demonstration at a vision show or supplier
shop, all you really know is that the vision sys-
tem is able to locate features on that particular
rack and provide osets to a robot to load a
small sample of parts into the rack.
A fully automated cell has to handle all
the variation in parts and racks that you
will see in production and anything else
you can throw at it. A successful applica-
tion has planned for and tested the varia-
tion before installation. Te process steps
we used at Ford Motor Co. for one such
application are
Dene system requirements
Develop vision solution feasibility
Optimize overall system process
Develop hardware/software architecture
Vision solution development
Integration & debug
Validation
Installation
Te process ow for the overall cell op-
eration should be dened. It has to include
both dened functions and any possible
exceptions to the ow. For example, you
might want to design the cell to start load-
ing only completely empty racks. What will
happen if the driver drops o a partially
loaded rack? How will the automation
identify the situation, and what is the next
step? Any interaction that requires operator
intervention should be closely analyzed.
Since you are developing an automated
cell, an operator might not be readily avail-
able to x the situation and resume cell oper-
ation. You could lose substantial throughput
if the cell sits in an error state for an extended
period. Also, there are safety concerns if the
operator is expected to enter the area around
the robot or the drop o area where the truck
drives, or if the fork lift driver has to exit his
vehicle to tend to the rack or automation.
Before you get too far down the path,
you need to verify the feasibility of us-
ing vision to solve this application. At this
point, you might want to consider both ro-
bot-mounted and stationary cameras. Eval-
uate the robustness, cost, and complexity
of each solution. Te shipping rack should
be evaluated for strong, repeatable features
that can be used to provide osets.
Dierent solutions using competing sup-
pliers systems and various lighting tech-
niques can be analyzed. Determine the
camera resolution requirements. Te solu-
tion with the highest probability of success
at an aordable cost should be selected
not necessarily the lowest cost proposal.
Once you understand the issues with the
vision application and have selected a vision
integrator, it is time for the team to discuss
the entire system. An optimal system solution
requires trade-os between the vision, robot,
material handling, cell controller, and other
automation devices in the cell. By knowing
the hardware and software capabilities of the
selected equipment, you can determine which
equipment handles which tasks.
Te entire project team, including vision
integrator, process equipment integrator,
plant operations, material-handling spe-
cialists, and system integrator should meet
early and often to develop the concepts and
architecture for the overall solution. Repeat
steps 1 through 4 iteratively until the team
is satised that the solution meets the cost,
complexity, and robustness targets identi-
ed. Decisions made at this point in the
project will determine the overall success
you will achieve.
Once the architecture is dened, an over-
all system specication must be written. Be
specic. Missing details (that is, just say-
ing that the robot must automatically load
parts) can lead to disappointments with
the operator interface, calibration proce-
dure, and packaging. If the vision solution
for robot guidance is a commercial prod-
ucteither embedded in the robot control-
ler or a third party PC-based vision sys-
temyou have an advantage because you
can evaluate the system before you make
the decision to purchase. When each sub-
system has a specication for its tasks, the
vision integrator can then design the spe-
cic solution.
Do not underestimate the amount of en-
gineering it takes to provide a good operator
interface and calibration procedure. Many
suppliers believe their intellectual proper-
ty is contained in their vision algorithms.
In practice, customer satisfaction is deter-
mined almost entirely on ease of recovering
from problems during production. Vision
algorithm performance is simply the mini-
mum requirement to do the application.
Te integration, debug, and installation
procedures follow standard project man-
agement guidelines. While they are also
critical to the success of the project, it is
not a subject covered in this article.
WHAT WE DID
We shipped a rack and some parts to the vi-
sion integrator to program a feasibility dem-
FIGURE 2. Target ducials were drilled in the rack for the vision system to use for
hook location (left). The vision system locates the black targets against the white
background (right).
0712VSD_26 26 11/29/07 11:03:07 AM
Allied Vision Technologies has reached another milestone in the digital image-processing
with its newest family of cameras: the Stingray. Designed as an innovative transformer
camera, it adapts to every conceivable application and offers a wide range of variations
with an unbeatable price / performance ratio, intelligent image-processing functions,
optimal networking capability, superior image quality and, besides the AVT modular
concept, the fastest IEEE1394b interface as well. But what else would you expect from
the worlds leading producer of digital FireWire cameras? www.alliedvisiontec.com
Your Fortune... The future of Digital Cameras...
S E E I N G I S B E L I E V I N G
0712VSD_27 27 11/29/07 11:03:39 AM
CAPTURE THE POWER OF BLUE
sensors | cameras | frame grabbers | processors | software | vision solutions
0712VSD_28 28 11/29/07 11:04:18 AM
Capture the Power of BlueCapture the Power of DALSA
Download the white paper 4 Keys to Maximizing Digital Image Capture | www.dalsa.com/4keys
www.dalsa.com
IMAGE SENSORS
DALSA designs and manufactures the highest-
performing CCD and CMOS image sensors
available, with uncompromised performance in high
sensitivity TDI line scan and two-dimensional area
scan technology.
www.dalsa.com/mv/sensors
CAMERAS
DALSA cameras offer unmatched combinations of
speed, responsivity, and dynamic range, resulting
in unrivaled image quality. Choose line or area scan
technology, proprietary CCD or CMOS sensors, and
Camera Link
Cables
- FireWire
Cables
gore.com/visioncable
Camera Link is a registered mark of the Automated Imaging Association (AIA).
FireWire is a trademark of Apple Computer, Inc., registered in the U.S. and
other countries.
w w w. v i s i o n - s y s t e m s . c o m Vi s i o n S y s t e ms De s i g n De c e mb e r 2 0 0 7 31
IndustrialAutomation Products
Ethernet switch mounted in the junction
box further consolidated the wiring. And
since the smart camera could trigger the
LEDs with on-board I/O, we only had to
route two cables on the robot itselfan
Ethernet cable and power.
Te cable was further sectioned with a
junction at the base of the robot, so that
if the robot-mounted cable broke an op-
erator would not have to go into the cable
trays and sh out the wires. Tis greatly re-
duces maintenance time. Compare this to
a PC-based vision system and all the wires
that would have to be routed, and you can
see why people get so excited about smart
camera packaging for machine vision. We
paid a $15,000 premium to use smart cam-
eras over PC-based vision for this applica-
tion, but everyone thought it was worth the
money (see Figs. 3 and 4).
Te cell integrator designed the robot
end-eector tooling and did a good job of
protecting the cameras from a robot crash.
Because the smart camera comes in a fac-
tory hardened enclosure, it did not have to
be put in a separate enclosure, saving both
money and weight. A screw-on lens protec-
tor cap was used to cover the lens. Te in-
tegrator provided a heavy gauge guarding
around the cameras and lights. Near-in-
frared LEDs were selected for the lighting
to protect the fork truck drivers from the
annoyance of ashing lights. Tese devices
are strobed and packaged in a factory-hard-
ened, thermally e cient housing, ensuring
their 50,000-hour-rated lifetime exceeds
the anticipated life of the cell (see Fig. 5).
If you talk about the robot crashing, ro-
bot suppliers will tell you that robots dont
crashthey are reliable and the controllers
dont get out of sequence or go in the weeds.
While it might be true that robots dont
crash, people do crash robots. Most crashes
happen during or after some maintenance
operation. We had two crashes already on
this robot, one during system debug and
another when there was a damaged rack in
the cell and the maintenance operator, at-
tempting to remove the part, got the coor-
dinate frames mixed up. Te cameras sur-
vived both crashes (see Fig. 6).
Even with smart cameras, the vision sys-
tem requires a PC to provide an operator
interface and do the math calculations of
3-D stereo vision. Te PC can also be used
to store images of failed operations and data
logging. Dont forget to make the PC ro-
bust, since it is required for the operation
of the cell. A desktop PC in an enclosure is
not an industrial PC. You also need to back
up the hard drive, either over the network
or with a redundant hard drive.
Te programming console for the vision
guidance system should be designed for an
operator, not vision engineer. While engi-
FIGURE 5. The smart camera was well
protected from a robot crash and, since it
came in a factory hardened shell, did not
require a separate enclosure.
FIGURE 6. Two cameras with IR LED light
view the hooks to nd the full depth of
eld position of the hook in the rack
using stereo vision. This geometry is
used to guide the side sill to the back of
the arm without scraping.
0712VSD_31 31 11/29/07 11:05:50 AM
De c e mb e r 2 0 0 7 Vi s i o n S y s t e ms De s i g n w w w. v i s i o n - s y s t e ms . c o m 32
IndustrialAutomation Products
neers get excited about programming vision
systems, operators and skilled trades can be
intimidated. Operators should not have to use
gray-scale or pixels. Osets are in millimeters.
Make it simple and graphical for ease of use.
HANDLING THE SYSTEM
You will need functions available to handle
vision system setup when things go wrong
in the cell. A really important function is
calibration verication. When the robot
can no longer load parts in the right loca-
tion, the rst thing people suspect is what
is considered to be the most complex thing
in the cellthe vision system.
Tey will tweak the calibration between
the vision system and robot, and usually
end up making the problem worse. If you
provide a calibration verication function,
the operator can test calibration before
changing it. One simple push of a button,
and the system will say whether the cali-
bration is OK or not. If it is OK, they can
look elsewhere for problems.
Another important function is autocalibra-
tion. If a calibration target is in the cell, anoth-
er push of the button, and the system can reca-
librate itself without the operator ever having
to lock out the cell and enter the work area.
Te third function is simplied camera
replacement. If you provide graphic over-
lays, you can guide the repositioning of a
camera so that the system will require only
minimal recalibration. A post in the cell is
our calibration target (see Fig. 7).
Te vision system consists of a PC and
three smart cameras running under a com-
mercial robot-guidance software package. It
is a slave device to the robot controller that
gets its commands from the cell controller
PLC. All communication is done over Eth-
ernet. Te three smart cameras are mounted
on the robot tooling: two cameras at the top
form a stereo pair to nd the hook position
and angle and the bottom camera locates
the center between the guides.
Te vision PC calculates osets and angles
from the location of target ducials on the
front tab and the back of the rack. Te front
tab establishes the position of the hook so
that the robot can center the hole in the part
over the tabsimilar to threading a nee-
dle. Te rear ducial in combination with
the front ducial establishes the angle of the
hook. Once the rst part is loaded, it is as-
sumed that the angle of the hook does not
change for the next six parts on that hook,
but the location is measured again for each
subsequent part in case the rack was moved
between parts. Camera 3 is used to center
the 6-ft part between the guide rails at the
bottom and the hook at the top.
Te cameras also use Ethernet to commu-
nicate with the vision PC, but their commu-
nication is local and restricted to the vision
PC in the cell. Te robot controls all motion
and positions within the cell and only uses
the vision system to provide osets to the
load positions. Using robot-mounted cam-
eras enables a lower-risk, more-robust solu-
tion because the magnitude of allowable o-
set for robot motion can be restricted to what
is reasonable for an undamaged rack. If the
vision camera should misidentify a feature
and make an erroneous calculation, an error
would be agged before the robot was sent
to an invalid location.
IN OPERATION
Te vision PC initiates an acquire-image
command to the smart cameras. Te cam-
eras snap the picture, process the image, and
return the location of the target it identi-
es. Te intelligence of the system resides
in the PC, which under the control of the
robot guidance software package, uses the
target locations to perform one of its speci-
ed functions: calibrate, verify calibration,
and provide the robot with osets and angles
for loading parts on the hooks.
Before loading the rst part in the rack
it is necessary to check for the shipping
bar to be in the down position and for
the rack to be empty. Te team decided
not to automatically load partially lled
racks. Tis decision simplied the system
design considerably, and t in with stan-
dard production practice. Initially, it was
going to be the vision integrators respon-
sibility to verify that the rack was ready
to load. Vision cameras were considered,
but these two checks were actually quite
di cult to do with vision.
We decided to use a laser safety curtain
to verify empty racks and a laser proxim-
ity switch for the shipping bar position.
Breaking a through-beam safety curtain
is intuitively more straight forward than
programming a vision system to look
through an empty rack onto the factory
oor and verifying no parts are in the rack.
Te system integrator took responsibility
for these two checks, again simplifying
the overall system.
Tis project was successful primarily be-
cause the entire team got together early and
made good trade-os in the system design.
What could have been a di cult vision so-
lution turned out to be almost trivial for the
vision system: nding three black circles
against a white background. Te most ro-
bust automation solutions are those where
no detail is overlooked and good engineer-
ing principles are used to solve complex
problems with the simplest approach.
KUKA Flexible Production Systems
Sterling Heights, MI, USA
www.kukarobotics.com
Sha, Brighton, MI, USA
www.shainc.com
CEC Controls, Wixom, MI, USA
www.ceccontrols.com
C
o
m
p
a
n
y
I
n
f
o
FIGURE 7. A calibration post is positioned
in the cell, and trained to the robot
coordinate system during installation.
After the installation, the column post is
used to verify calibration and to recalibrate
the system in case of camera replacement
or robot crash.
0712VSD_32 32 11/29/07 11:06:37 AM
0712VSD_33 33 11/29/07 11:07:09 AM
0712VSD_34 34 11/29/07 11:08:41 AM
P
r
o
f
i
l
e
www. v i s i o n - s y s t e ms . c o m Vi s i o n S y s t e ms De s i g n De c e mb e r 2 0 0 7 35
Industry Solutions
By Wi nn Hardi n, Contributing Editor
We use true optical character verication, which compares fonts to libraries of digital images,
explains Systech software engineer Mike Soborski. We use digital stacking to create variet-
ies of the code that include common skew and other distortions based on our knowledge of
what typically happens in a pharmaceutical laser printer. Unlike items in the food industry,
a vial of medicine may cost $100 and be produced at 350 vials a minute. Pharmaceutical
manufacturers do not want this line to go down. Engineers spending their time tweaking
thresholds or optical character recognition tools is simply not acceptable.
OCV supports FDA-compliant
packaging line
Multiple machine-vision stations inspect and verify
pharmaceutical vial-packaging system.
A
major pharmaceutical manufacturer with a
packaging facility in Puerto Rico recently
asked Systech International to design, qual-
ify, and install a vial-packaging line compli-
ant with US FDA 21 CFR standards, including label
application, tray lling, and palletizing. Te line had
to be exible enough to accommodate new products
or to add new sensors and software and to report ca-
pabilities without having to be recertied.
Any software or hardware change to a pharmaceu-
tical manufacturing line must provide an audit trail
for 100% of lot production. If a systemparticularly
softwareis not properly designed and compartmen-
talized, changes to any part of the system can warrant
a revalidation of the entire line. By using o-the-shelf
modular software proven to comply with 21 CFR 211,
only the new modules and aected supervisory func-
tions must be validated when something is added to
or subtracted from the production line.
Te Systech manufacturing line includes two mono-
chrome optical-character-verication
(OCV) machine-vision inspection sta-
tions at the front-end cut-away labeler
and back-end expiration-lot labeler, a
third for Data Matrix verication, and
a fourth color vision system to check col-
or codes on the vials in packing trays be-
fore shrink wrapping, casing, and pallet-
izing. Each vision station provides a key
inspection and documentation step on
the way to FDA compliance.
SINGLE POINT OF CONTROL
An end-to-end audit trial for pharmaceutical produc-
tion and packaging is simplied if the line has a single
point of operation and data I/O for all but operators
with management clearance and above, according to
Systech International regional sales manager Len Va-
leo. If the system has only one data I/O and control
point, the likelihood of operator error altering a sub-
system or resulting in lost data is greatly reduced.
Supervisory-control-and-data-acquisition systems
for industrial manufacturing and automation inher-
ently do not provide secure access or data tracking to 21
CFR levels. About 75% of all pharmaceutical packag-
ing equipment is not compliant with 21 CFR Part 11,
says Systech International regional sales manager Len
Valeo. Placing our Advisor Line Management Soft-
ware on top of the Allen-Bradley PLC network enables
the equipment with a form of compliance.
A secondary proprietary physical layer based on a CAN
network connects and secures all intelligent equipment
features, advantages, benets
0712VSD_35 35 11/29/07 11:09:32 AM
Barcode 1
tray
ADVISOR display
(tray packer)
ADVISOR display
(tray packer)
Camera
Data Matrix
Camera
OCV
Camera
OCV
Color
camera
vial count
Barcode 2
insert Barcode 3
insert
To shipping
Master
5/05 PLC
ISX
computer
Hub
RS-232
RS-232
Labeler
laser printer
Tray packer
laser printer
Conveyor
belt
Ethernet
Cut-away
labeler
(with PLC)
Expiration-lot
labeler
(with PLC)
Tray packer
(with PLC)
Discrete
I/O
Discrete
I/O
Shrink wrapper
(with PLC)
Case packer
(with PLC)
TIPS ADVISOR panel
De c e mb e r 2 0 0 7 Vi s i o n S y s t e ms De s i g n w w w. v i s i o n - s y s t e ms . c o m 36
ProleIn Vision Solutions
on the line through a Systech TIPS LINK
proprietary cable and Vision Integration Kit
(VIK) I/O box attached to the PLC. Te VIK
provides a high-bandwidth data channel to
the vision system, as well as access to the PLC
through an embedded Allan-Bradley indus-
trial Ethernet-compatible embedded chipset.
Te high-bandwidth channel allows for low-
latency transmission of images from all three
machine-vision inspection stations to the Sys-
tech central ISX computer, which runs Sentri
vision software and associated image-process-
ing modules along with the Advisor super-
visory program and the main HMI for the
pharmaceutical production line. Te ISX is
an industrial, Pentium-based PC with VIK
master card to access the Systech network and
Allan-Bradley SLC card to access the indus-
trial Ethernet PLC network.
To initiate a lot run, a supervisory-level
operator selects the production lot recipe
from a library of recipes stored on the ISX
central computer. Te Advisor software
queries the operator for each relevant vari-
able (for example, lot size, expiration dates)
and initiates the production run. Advisor
sends out operational information to the
cut-out labeler, expiration/lot labeler, tray-
loading machine, smart barcode readers,
and associated vision inspection stations,
as well as resetting counters and relevant
settings for each PLC on the Allan-Brad-
ley network (see Fig. 1).
Te labeler and tray PLCs both have laser
printers, while the lot/expiration code print-
er uses a Lastec laser etching system, ex-
plains Valeo. If Advisor wasnt sending the
information to these printers, the operator
would have to manually input the informa-
tion separately into each
machine, increasing the chances of errors.
Valeo says customers can save upward of
$30,000 by eliminating the need for a com-
plete barcode reader system and HMI and
by using $5000 smart barcode readers from
Microscan or Accu-Sort directly integrated
to the Advisor program and HMI. Once
the operator initiates the lot, Advisor sends
a lock code to every intelligent machine
on the system so that variables cannot be
changed locallywith few exceptions.
SECURING THE UNSECURED
Te majority of packaging equipment used
in pharmaceuticals is not 21 CFR compli-
ant, which means the entire line is not FDA
compliant. Te cut-out labeler PLC, expira-
tion/lot-code printer PLC, tray PLC, and
shrink-wrap-palletizer PLCs are all con-
nected to the Allen-Bradley industrial
Ethernet network, but control is not se-
cure against local access. Recipes for each
machines operation are stored locally on
the PLC, which does not include the se-
curity levels required to ensure production
lot integrity based on 21 CFR standards.
However, the PLC can be programmed for
a variety of alarm codes.
Systech set aside a 1-bit alarm code based
on data accessed at the PLC. If the local
recipe is accessed, the alarm bit is set to its
highest level of 3. Te Systech system has
full access to each PLCs data via its con-
nection to the
Al l a n- Br a d-
ley industrial
Ethernet LAN.
When a Level
3 code is issued,
Advisor halts the line. It can only be re-
started by a management-level supervisor.
So while the actual changes to the recipe
cannot be guarded on the PLC, Advisor
does track who approved restarting the line
in response to a Level 3 PLC recipe access
event, providing the requisite audit trail
and making unsecured PLCs secure to 21
CFR levels (see Fig. 2).
KEY PRODUCTION STEPS
After the operator initiates the operation
and inputs all values and each PLC and
vision system is programmed with the ap-
propriate procedure based on the stored
recipe in the Systech program, lled glass
vials enter the labeler station from an ac-
cumulator. A laser printer inside the la-
beler machine begins printing the cut-out
labels with the appropriate information,
and an image of each label is captured by
a Systech CCD-610 monochrome digi-
tal camera through a local trigger. Te
camera is outtted with standard optics
and bandpass lter to reduce interference
from ambient light and focus on the il-
lumination from a nearby Systech LED-
RR01 red LED ringlight. Te image is
transferred back through the VIK box to
the VIK master at the ISX host PC and
then passed into the PC memory where
the Systech OCV tool takes over.
If the label is passed by the Sentri in-
FIGURE 1. The Systech
TIPS Advisor network
for pharmaceutical
production lies on top
of the Allan-Bradley
Ethernet network
and uses minimal
proprietary discreet
runs, as well as the
AB network, to
communicate, initiate,
secure, and control
all equipment on the
packaging line.
0712VSD_36 36 11/29/07 11:10:03 AM
.BDIJOF7JTJPO-JHIUJOHT
The groundbreaking PHLOX
technology sets apart from the others
Uniformity : up to 5 %on the
backlight
Luminance : up to 50 000 cd/m2
Heat free : a maximum
of 20C/68F, above room
temperature for all our standard
products, ensuring a longer
lifetime
Ultra slim : 8 mm - 0,315 thick
High protection : IP 65
Large sized (up to 600x600 mm)
and custom sized products also
available at low cost and fast
delivery
Energy saving and better
efficiency
Further Benelts :
24 months warranty
Certificate of uniformity and
luminance provided with each
product
Seal of quality on each product
www.avisionahead.com
info@phlox-gc.com
1)-09.BDIJOF7JTJPO-JHIUJOHT
"7*4*0/
")&"%
w w w. v i s i o n - s y s t e m s . c o m Vi s i o n S y s t e ms De s i g n De c e mb e r 2 0 0 7 37
ProleIn Industry Solutions
spection software, the labeler PLC ex-
tracts the label from the printed roll and
attaches it to the vial. A second printer
and monochrome vision system verify a
Data Matrix code, which is then also ap-
plied to the vial. Te vial passes through the
labeler, and a smart barcode reader veries
a separate Code 128 dense barcode on the
label against the programmed recipe. Us-
ing multiple codes provides redundancy in
the operation.
Te vial then enters the lot-code/expi-
ration-date laser printer, which etches the
lot code into the glass in the cut-out area.
A second OCV system veries the lot code
and expiration date (see Fig. 3).
Te vial continues to the tray packer
PLC. A pair of barcode readers read the
Code 128 codes on the plastic tray and
plastic insert and record those data with the
lot information on the local PLC, which
is then passed to Advisor. Te PLC places
vials in each of the trays holes and
places a colored dot on the top of
the vial that varies by recipe. When
the tray is lled, a Systech CCD-
645 megapixel color CCD camera
captures an image of the entire tray
from above and uses color thresh-
olding to identify each colored dot.
Te dots are counted to verify the
tray is full and checked against the
color code included in the recipe.
Systech has a variety of color soft-
ware modules for its Sentri vision
systems for pharmaceutical produc-
tion. One interesting feature is the
ability to do 15-bit digital stacking
of each color to identify all RGB
color space values that are near to
the target color code. Te color tool
creates a look-up table for each dot
based on nearby color extrapola-
tion. Tis allows the system to ac-
commodate slight changes in the
dots color without falsely rejecting
the vial. Also, the tool generates a
3-D RGB color space cub that can
show all colored labels used for lot
tracking. Tis allows the engineer
to verify that no colored dot can
be mistaken for another dot to its
proximity in color space.
After the color images are sent
back via VIK channel to the ISX comput-
er and passed, the Systech system sends a
signal for the PLC to pass the lled tray to
the shrink wrapper and robotic palletizer
for shipping to the customer.
FIGURE 3. Systech software for optical
character verication allows the system to read
slightly distorted or skewed codes.
FIGURE 2. During 21 CFR testing to validate
pharmaceutical production, the TIPs Advisor
screen shows the operator a view of a factory
acceptance test of vial-inspection packaging line.
Accu-Sort Systems, Telford, PA, USA
www.accusort.com
Lasetec, Torrance, CA, USA
www.ltec.com
Microscan, Renton, WA, USA
www.microscan.com
Rockwell Automation (Allen-Brad-
ley), Milwaukee, WI, USA www.
rockwellautomation.com
Systech International, Cranbury,
NJ, USA
www.systech-tips.com
US Food & Drug Administration,
Rockville, MD, USA
www.fda.gov
C
o
m
p
a
n
y
I
n
f
o
0712VSD_37 37 11/29/07 11:10:29 AM
We offer the broadest spectrum of Machine Vision systems for all
fields of application from 1D/2D code readers to color sensors to
smart cameras to PC-based image processors. Wherever your
Machine Vision application takes you, we can help get you there.
Improve quality and productivity with SIMATIC sensors
For more information or to contact a sales representative, call,
e-mail, or visit our web site.
2
0
0
7
S
i
e
m
e
n
s
E
n
e
r
g
y
&
A
u
t
o
m
a
t
i
o
n
,
I
n
c
.
Improve your quality with streamlined processes
Machine Vision made simple
simatic
Siemens Energy & Automation, Inc. 1-800-964-4114 ref. code VSD1107 info.sea@siemens.com www.sea.siemens.com
0712VSD_38 38 11/29/07 10:52:58 AM
w w w. v i s i o n - s y s t e ms . c o m Vi s i o n S y s t e ms De s i g n De c e mb e r 2 0 0 7 39
ProductFocus
S M A R T C A M E R A S
Sensors tackle machine-
vision applications
Vision sensors are closing the gap between traditional photoelectric
sensors and more complex image-processing systems.
By Andrew Wi lson, Editor
A
t Pack Expo in Las Vegas, NV,
USA (October 2007) this year,
one emerging trend that was ev-
ident was incorporating more
sophisticated machine-vision systems
into packaging and production systems.
But those expecting to see sophisticated
machine-vision systems integrating expen-
sive telecentric lenses, large-format cameras,
high-speed Camera Link frame grabbers,
and image-processing software would sure-
ly have been disappointed.
While many systems featured relatively
low-cost devices to read one- and two-di-
mensional (1-D and 2-D) barcodes, very
few used more costly components with
which to perform machine-vision tasks.
Te reason, it seems, is multifold. Although
many packaging-system designers would
like to incorporate machine vision into
their systems, the expense of developing
and deploying such systems is prohibitive.
What developers are looking for is an inex-
pensive way to rapidly deploy single products
that perform simple functions such as pres-
ence detection, color analysis, and barcode
reading. Familiar with using low-cost pro-
grammable logic controllers (PLCs), system
integrators are now exploring the uses of smart
vision sensors to perform these functions with-
out the need to understand how to develop or
deploy machine-vision algorithms.
Of course, this trend is not new. One of
the rst companies to recognize the demand
for a low-cost way to solve relatively simple
machine-vision tasks was DVT, now part
of Cognex. Rather than develop simple bar-
code readers, DVTs concept was to develop
FIGURE 1. Range of smart sensors now available differ in the types of imaging
sensors used, the speed at which they operate, and the software support they
offer: (a) Cognex Checker 200 Series,(b) Datalogic Matrix 400, (c) Banners
PresencePLUS, (d) NI 1722, (e) Balluff Sharpshooter, and (f) Baumer Optronics
VeriSens sensors.
a)
c) d)
b)
e)
f)
0712VSD_39 39 11/29/07 10:53:33 AM
VGA resolution
(640 x 480 pixels)
De c e mb e r 2 0 0 7 Vi s i o n S y s t e ms De s i g n w w w. v i s i o n - s y s t e ms . c o m 40
ProductFocus on Smart Cameras
relatively inexpensive, compact, rug-
ged vision sensors that incorporated
the functions of a smart camera and
PLC. Tese include an integrated im-
ager, memory, processor, Ethernet, and
digital I/O capability, along with the
companys Intellect software that uses
an intuitive user interface and a library
of easy-to-congure vision tools.
For the packaging and production
industry, ease of mounting, industri-
al connectors, and integrated lighting
of smart sensors provided a way to
migrate from the limitations of pho-
toelectric sensors to provide addition-
al functionality with a moderate cost
increase. It is a concept that has been
further extended by Cognex in its lat-
est Checker 200 series (see Vision Sys-
tems Design, June 2007, p. 22).
NEW PRODUCTS EMERGE
Realizing the success of these products,
other companies are now developing
similar smart vision sensors with dier-
ent levels of sophistication (see table). In-
terestingly, these companies are emerging
from existing manufacturers of low-cost
barcode readers, companies known for
industrial control products, and existing
machine-vision vendors. While many of-
fer smart vision sensors rmly targeted to-
ward 1- and 2-D barcode reading, more
sophisticated machine-vision functions
are slowly being added to these products.
One thing is apparent when compar-
ing these products. Many of the techni-
cal hardware specications are very simi-
lar. Nearly all of the smart sensors on the
market today oer integrated sensors, pro-
cessors, memory, digital I/O, networking
capability, and LED lighting. Even more
noticeable, the design of such products is
also strikingly similar to products origi-
nally developed by DVT (see Fig. 1).
Te choice of which smart sensor to
use is rmly application-dependent. Te
sensor size, frame rate, type of built-in
FIGURE 2. Smart sensors with 1280 1024
resolution deliver a larger eld of view without
sacricing resolution. Microscan MS-4 EZ Match
can read barcoded test tubes or combinations of
linear and 2-D symbols while capturing the height
of the test tube and its cap for inspection
0712VSD_40 40 11/29/07 10:54:26 AM
www.stockeryale.com
MORE
POWE R
Uniform laser line generator
1 W (infrared); 500 mW (red)
Focusable
High pointing stability
Operating temperature:
-20C to +55C
Lasiris
PowerLine Laser
StockerYale Inc.
Tel.: (514) 685-1005
Fax: (514) 685-3307
1-800-814-9552
lasers@stockeryale.com
NASDAQ: STKR
Copyright