Introduction to Machine Vision

Introduction to Machine Vision

Table of contents

Introduction to Types of machine Rule-based vs.

machine vision vision systems and AI-powered machine
what they can do vision

PAGE 2 PAGE 5 PAGE 9

Machine vision The importance of Image analysis

hardware components lighting in machine software
vision applications

PAGE 13 PAGE 16 PAGE 18

What is machine vision?
Machine vision is an essential component of how digital systems interact with the real
world. It lets automated systems see components, products, patterns, codes, or other
objects and use that information to make decisions.

Since it allows manufactured parts and products to be inspected, measured, and sorted,
machine vision has vastly increased the power and flexibility of industrial automation.
Machine vision does all this at high speeds and high accuracy, improving product quality
and reducing waste.

Machine vision also lets automated equipment locate objects, identify them, and save
information about their material, condition, orientation, and other details for later analysis.
That data is critical to factories looking for efficiency gains.

And every package that makes its way through today’s automated logistics
warehouses does so with the help of barcodes that are read by machine vision. Those
codes are used to track and identify packages throughout their journeys, making sure
they get to their intended destinations.

All of these activities are powered by machine vision:

cameras that let computers interpret real world objects.
You may also hear products in this category referred to as
“computer vision,” which is a broader term and sometimes
Machine vision lets automated systems see
describes more theoretical research systems that analyze components, products, patterns, codes,
images. Machine vision products, on the other hand, are
practical, connected to industrial automation systems in or other objects and use that information to:
factories and warehouses.
▪ Make decisions at high speeds and high accuracy, improving product quality
and reducing waste.

▪ Generate data that is critical to factories looking for efficiency gains.

Introduction to Machine Vision 2

Who uses machine vision?
Machine vision is constantly evolving, and performs an ever-growing range of tasks in two main types of organizations:

Manufacturing facilities and factories Logistics providers and warehouses

use machine vision to automate operations throughout the use machine vision to automate routing, tracking, and shipping, typically by
manufacturing process. reading barcodes or other encoding symbols.

Introduction to Machine Vision 3

What are the benefits of machine vision?
Because it can see with better resolution, across a wider spectrum, and process tasks faster than human vision, machine vision provides a wide range of benefits:

Improve product quality Overcome labor shortages

Products rejected for defects are a significant source of Manufacturing is facing significant labor shortages, and the
cost, waste, and reputational damage. Automated inspection problem is expected to get worse. The work can be tiring and
with machine vision improves speed and accuracy, catching difficult. By automating manufacturing with machine vision,
problems of many kinds before they’re packaged or shipped skilled labor can be reserved for higher-value activities.
and allowing human inspectors to be reserved for an
increasingly small number of difficult cases.

Trace parts and products Improve processes

By reading codes on products and packages at every step Machine vision instantly detects changes in product quality
from initial production to shipping to final sale, machine and keeps a visual record of every step in a product
vision systems can provide critical tracking information. This lifecycle. This form of big data reveals process bottlenecks,
lets shippers know their current location, quickly detect any declining machine function, and common sources of error,
delays or shipping errors, and trace any damage or other making continuous process improvement possible.
problem back to its source.

Increase productivity/overall equipment Ensure compliance

effectiveness The machine vision-generated data and images used to
Machine vision systems speed up operations and make process improvement decisions also provide the data
decrease cycle times, and their performance doesn’t needed to comply with reporting regulations in industries
deteriorate over the course of a shift. They provide the like pharmaceuticals, medical devices, automotive, food and
information to make the most efficient use of every piece of beverage, and more.
equipment on the floor.

Reduce waste Improve safety

By catching manufacturing flaws, identifying overfill, or With industrial automation made possible by machine vision,
pinpointing the causes of defects, machine vision can reduce workers can minimize their interactions with noisy and
waste and scrap rates in multiple ways. Over time, this can potentially dangerous machinery. If there is an equipment
help control overhead and bring down raw materials costs. problem, they can often diagnose and adjust it remotely,
through the information machine vision provides.

Introduction to Machine Vision 4

Types of machine vision systems and what they can do
While machine vision systems offer a wide range of features and options, there are three main categories to consider.

Line scan Area scan 3D vision

Line-scan cameras are used in continuous inspection Most machine vision uses 2D cameras, also called area 3D vision systems add depth to their images, sometimes
applications such as web manufacturing. They take a scan systems. These can be simple, single-purpose using lasers to measure distances and compute depth.
wide but very thin image, typically as material moves sensors or more full-featured systems. Sensors are The addition of depth can increase complexity and cost
past the scan area, and use software to reconstruct the lower cost, smaller, easier to deploy, and often more to some degree, but that additional information is critical
image line by line. rugged than cameras with a wider range of features. to some applications.
Generally, full-featured systems are more configurable,
They are significantly faster in these applications than perform more complicated tasks, and can take larger For example, 3D vision is essential in guiding a robot
standard 2D cameras. Examples include inspecting and higher resolution images. arm to precisely reach out to grab an object in the
fabric, paper, and other soft goods. correct orientation, no matter where it is in space. 3D
The decision between sensors and more complex vision is also used to automate difficult cutting and
machine vision systems comes down to the task being welding processes.
performed, the format of the data output, cost, and ease
of use.

Introduction to Machine Vision 5

What does machine vision do? Defect detection
Defects can occur in any part of a manufacturing process, from problems with the quality of raw materials or parts
Machine vision providers sometimes classify what through final inspection. Inspections to catch defects have traditionally been done manually, by trained workers.
their systems can do with the acronym GIGI, for
Guidance, Identification, Gauging, and Inspection. Machine vision is a significant improvement over human inspection: it operates at production line speeds, doesn’t get
This guide breaks down the types of applications tired, can detect even small and unexpected defects, and stores information for continuous operational improvement.
into slightly more detail:
EXAMPLES:
▪ Defect detection

▪ Object detection and counting

▪ Measuring/gauging

▪ Locating/guiding/positioning Hair

▪ Barcode reading
Looking for loose connectors, poorly Detecting flawed photovoltaic cells in solar Finding contaminants or other problems in
▪ OCR/OCV
soldered wires, bad seams, or improperly panels or defects in semiconductor wafers food products
crimped tubes or EV battery assemblies

Object detection and counting

Determining the presence or absence and counting objects is a widely used function in inventory management, on
production lines, and before releasing or accepting shipments. Both manual inspection and mechanical counting are
slow and prone to error, compared to high-speed and consistently accurate machine vision systems.

EXAMPLES:

Confirming the presence of all electronic Verifying the presence of components Counting products in a package or on a
components on a printed circuit board such as clips, screws, springs, labels, pallet
seals, manuals, inserts, or accessories

Introduction to Machine Vision 6

Measuring/gauging Locating/guiding/positioning
Precise manufacturing requires accurate measurement of distances, areas, Functions such as assembly, pick-and-place, and inspection depend on machine
diameters, and more. Manual measurement with gauges, calipers, or inspection jigs vision’s ability to locate a part, whether on a conveyor or in a bin. Machine vision
is slow and introduces inconsistencies. guides robots in picking and placing parts and ensures accurate micron-level
positioning in precision assembly.
Machine vision consistently and accurately measures down to the micron level, while
parts are on the line, at high speeds. Each image and its associated data can be If a part is out of place, machine vision measures the difference between
stored in case of warranty or compliance issues. desired and actual location and orientation and communicates that to a robot or
programmable logic controller (PLC) to realign the part.

EXAMPLES: EXAMPLES:

Capturing the dimensions of cast and Measuring the roundness and angle of tips Locating parts on a conveyor for inspection Guiding robots on automated automotive
injection molded parts on parts assembly lines

Determining label positions or package sizes Assembling microchips that require micron-
level precision

Introduction to Machine Vision 7

Barcode reading Reading text: OCR/OCV
Barcodes are used to track and identify raw materials and finished goods throughout Barcodes are everywhere in modern industry, but human-readable printed text is
the supply chain. They include 1D and 2D codes, such as UPC codes on retail still essential, particularly for retail, manufacturing, pharmaceutical, and food and
products and data matrix codes on packages. beverage supply chains, to identify sell-by dates, lot numbers, and other important
information. For this text to be useful in modern high-speed processing, it must be
Barcode readers, process information at high speeds, and correctly read codes that reliably readable by machines. Machine vision systems can read these codes in
are partially torn, obscured, smeared, or distorted. Image-based barcode scanners milliseconds with 99.99% accuracy.
collect images of the barcodes they read, allowing for analysis of no-reads or
misreads to diagnose problems, such as a clogged print head, damaged codes, or Optical character recognition (OCR) and optical character verification (OCV) both
inadequate lighting. Multiple camera systems can scan from multiple sides of a part or identify and interpret text in images, with one key difference. OCR reads text to trigger
package at once to increase read rates when not all codes are oriented the same way. another action, while OCV is used to verify the quality of text against a known standard.

EXAMPLES: EXAMPLES:

Tracking packages as they travel through a Ensuring the correct components are Classifying a part based on the characters Checking if a sell-by date or lot number is
logistics warehouse assembled printed on it (OCR) printed correctly (OCV)

Providing accurate traceability of medical Distinguishing authentic from counterfeit Establishing traceability to meet regulatory
supplies products (OCR) demands (OCR)

Introduction to Machine Vision 8

Rule-based vs.
Rule-based systems
Traditional rule-based machine vision uses specific
if-then rules to make decisions about an image. “If

AI-powered
the bottle is filled to at least this level, let it pass.” The
rules are programmed by vision engineers with deep
knowledge of the best way to get the desired output.

machine vision The rules are used to create programs that tackle specific
jobs, such as detecting the edge between light and dark
areas or measuring the distance between two points.
There are two ways machine vision can be used Rule-based systems
These rules can then be used one after another on an
to make decisions such as counting, classifying,
or approving and rejecting items. Rule-based
image, in what is called “tool chaining,” to perform a wide work on consistent parts
range of sophisticated tasks. They’re also very good at
systems follow user-programmed, step-by-
high-speed, high-accuracy inspections on products or Technical expertise: Vision expert with
step instructions to interpret images and make
parts that are predictable and consistent. programming knowledge
decisions. In contrast, artificial intelligence or
AI-powered systems use a database of reference
images to “learn” how to make decisions.

While rule-based machine vision is still the

prevalent technology, various types of AI-powered
machine learning have become capable and
flexible enough to take over in many applications.

Often, a combination of rule-based and AI-powered

machine learning can provide the most efficient
solution.

Often, a combination of rule-based and AI-powered machine learning

can provide the most efficient solution.

Introduction to Machine Vision 9

Deep learning
Deep learning uses AI to power machine vision systems by using examples to train the
software with labeled images until it can make distinctions on its own.

This training mimics how humans learn, helping the system learn how to make accurate
decisions without being thrown off by irrelevant variation.

Operators train deep learning systems with hundreds of labeled images, such as a
variety of possible defects in a manufactured part, or several different assemblies that
need to be classified.

One significant improvement over rule-based systems is that deep learning systems
learn to distinguish between a wide range of real defects and merely cosmetic variations
without needing to be shown every possible result — something time-consuming or even
impossible to accomplish using rule-based programming.

In contrast to rule-based systems, deep learning systems don’t require experienced

machine vision programmers or expertise to set up. They do need someone with good
Deep learning is designed for
knowledge of the products being inspected to provide hundreds of images of both complex applications
defective and good components.
Example images required: hundreds to thousands
Deep learning excels at assembly verification, defect detection, and classification of Total time: Hours to days
complex parts that vary unpredictably. It’s also good at providing accurate results even Technical expertise: Training in AI and programming
when parts are reflective or images are distorted.

Use case training: hundreds of example images Results: complex applications with significant variation

Introduction to Machine Vision 10

Edge learning
Edge learning is another type of AI, optimized to meet the needs of industrial
automation. It comes pre-trained to solve the types of problems typically faced by
industrial automation.

As a result of the pre-training, an edge learning system can be trained on a

specific industrial inspection problem with as few as five images, making it even
easier and faster to deploy than a deep learning system.

Edge learning enables line engineers to quickly implement an optimized

classification or defect detection function on their line, without specialized training.
Since it doesn’t require a sophisticated processor, it can be quickly deployed
where it is needed.

The advantages of edge learning lie in its speed, low demand on computing
resources, and flexibility, as well as its user-friendliness.
Edge learning is designed for
ease of use
Example images required: 5-10
Total time: Minutes
Technical expertise: No prior experience needed

Pre-loaded training and optimization Use case training: 5-10 examples Results: easy, fast

Introduction to Machine Vision 11

Comparing rule-based and Recommended uses for each type of system

AI-powered systems Rule-based vision Edge learning Deep learning

The decision between rule-based, deep learning,
and edge learning systems mainly comes down
to the type of application you’re implementing.

▪ Rule-based vision tools are effective at

a variety of specialized tasks with very
consistent targets, such as location,
measurement, and orientation. Complex defect detection
Measurement and gauging Classification
▪ Edge learning excels at repeatable tasks
with moderately consistent targets.

▪ Deep learning shines when looking for

complex defects or analyzing images with
significant variation.

Location and fixturing Assembly verification Tasks with significant variation

Robotic guidance Character reading Highly customized jobs

Introduction to Machine Vision 12

Machine vision hardware components

Choosing the right hardware Processor

for a machine vision system Embedded or in a separate PC

While most vision systems include a similar

collection of hardware, there are specific choices Image sensor
to be made for each component and they all Turns light into electrical signals
need to work seamlessly together.

The exact configuration of any machine vision Lens

system is dictated by the job requirements: by Collects and focuses light onto
starting with the application, you can work with a the sensor
vendor to determine the precise setup you need.

Cover
Protects the system from dust, water,
oil, and other environmental hazards

Introduction to Machine Vision 13

Image sensors Mono vs. color sensors
Machine vision image sensors are monochromatic,
At the center of any vision system, an image meaning each pixel only detects the intensity of the
sensor converts light energy into electrical light falling on it, not what color it is. They can capture
signals that can be analyzed by software. color images using red, green, and blue filters, but in the
vast majority of industrial applications, monochromatic
Image sensors are solid-state semiconductor imagery is a more effective choice. Since color filters
chips comprising millions of photodetectors reduce light intensity, they’re only used when color is
called pixels, which convert light into electrical absolutely required for a specific task.
signals. The two main technologies used are
charge-coupled device (CCD) or complementary
metal oxide semiconductor (CMOS). CMOS
technology is newer and taking an increasing Rolling vs. global shutters
share of the market, thanks to lower costs and Machine vision cameras use electronic shutters to
faster processing speeds, but CCD sensors are determine how the pixels are exposed to light. Despite
still used in applications that require extremely the use of the word “shutter,” as in an old mechanical
high-quality images. camera, the shutter is part of the way a sensor
processes the light falling on it, not something separate.

A rolling shutter image sensor exposes lines of pixels

sequentially. This kind of sensor is lower cost, but if the
target being imaged is moving rapidly, the image can A rolling shutter exposes A global shutter exposes
a single line of pixels at a all of its pixels at once.
be distorted. time.
A global shutter image sensor can expose all of its pixels
at once. They can be more complex to manufacture,
but prices have come down and they are widely used in
industrial applications.

Additional sensor features

Sensors range widely in a number of characteristics, including size, resolution, pixel size, frame rate, sensitivity, and
dynamic range. Each choice can affect other characteristics, as well as cost. Sensors can also have a wider spectral
range than the human eye and detect light into the infrared (IR), useful in some industrial applications.

There’s no single answer to “which sensor should I use?” Sensors should be chosen based on their specific
purpose, whether for a fast-moving assembly line, where a high frame rate might be essential, or detecting tiny details in
a complex product, where resolution and dynamic range might be more important.

Introduction to Machine Vision 14

Lenses Covers
Lenses collect and focus light onto A camera cover, also called a
image sensors. They are matched with housing or enclosure, protects
the sensor so they completely illuminate your expensive machine vision
the photosensitive area with a resolution camera from the dust, oil, water,
equal to or greater than the sensor’s food particles, impacts, and other
pixel size. hazards that come with operating
in an industrial environment.
Manual lenses need to be refocused if
the objects being observed change, or if Covers can also dissipate heat,
the camera moves. protect cable connections, and prevent tampering.

But this can be time-consuming if Covers are rated by how well they protect the camera, with IP67 being the
done frequently, or if there are a large usual rating for a dustproof camera in an industrial environment. An IP67 rating
number of cameras that need to be also means the enclosure is waterproof, so the operating area can be washed
adjusted. In these situations, autofocus down safely.
lenses are optimal.

Autofocus lenses adjust their focus automatically as needed, saving time if frequent
changes are expected.

Liquid lenses are a specialized subset of autofocus lenses that adjust their shape on the
fly to maintain sharp focus. They’re also smaller and more resistant to shock and vibration
Hardware
than mechanical autofocus lenses of similar spec.
requirements
Another factor in choosing the
type of processing is the required
hardware. Because of the amount
Liquid Lens Liquid Lens
of parallel processing it requires,
deep learning generally needs
to run sophisticated parallel
processors called GPUs. Edge learning requires less processing, and rules-
based algorithms the least.

Some industrial cameras have onboard or embedded processing. Such

cameras can typically do both rule-based and edge learning processing entirely
on their own. Deep learning may require a more powerful processor located in
a separate PC.

Introduction to Machine Vision 15

The importance of lighting in
machine vision applications
Poor lighting is the most common cause of poor machine vision performance.
Even sophisticated cameras and software can’t make up for inadequate lighting.

Good lighting for machine vision maximizes contrast on the features of interest
and minimizes it everywhere else. It also needs to be consistent; the lighting
setup should make sure normal variations in parts or their arrangement don’t
affect the contrast in the image.

Depending on the part’s texture (such as matte, reflective, or refractive) and its
shape, different lighting choices will change which portions of the image will be
brighter than others. Poorly chosen lighting might create glare, when light reflects
off a smooth part directly into the camera. A different position might occlude an
important spot or leave it in a shadow.

Some cameras come with integrated lights, which provide uniform illumination
that’s important when working with matte objects, among other uses. They’re
easy to deploy and work well in many situations.

Other conditions can require a separate lighting setup, which can involve
experimentation with different lighting angles and types to get the most effective
image. There’s no one best lighting setup for all machine vision applications.
Instead, lighting optimization depends on the application, specific requirements,
and the desired results.

Good lighting for machine vision maximizes contrast on the

features of interest and minimizes it everywhere else.

Introduction to Machine Vision 16

Types of machine vision lighting Filters
Filters can keep unnecessary ambient light
from entering a camera or increase the contrast
of a part or other object.

Backlighting Diffuse lighting

A flat, diffuse light behind the target forms a black Even lighting from multiple directions minimizes confusing
silhouette with maximum contrast and clear edges. reflections from a curved or shiny surface. Also called Color filters only let red, green, or blue light
cloudy-day lighting, diffuse lights can be domed, axial, or through. Depending on the objects being
flat, and help create an even background, making it easier examined, one color may provide higher
to detect features of interest. contrast.

Full/partial bright field lighting

Lighting an object from above creates strong contrast so
flat parts reflect into the camera.
Polarizing filters can eliminate glare from light
being reflected into a camera.
Multicolor/multispectral lighting
Using infrared (IR) lighting can make it easier to find
small surface defects within visually busy images, see
differences in material, and work in environments with
challenging visual conditions.

Dark field lighting

Shining a light from the side at a low angle (like a rising
or setting sun) helps show edges, height variations, IR and UV filters only let through light outside
texture, and contours. Background areas reflect light the visible spectrum and can often show
away from the camera, and appear dark, while the contrast differences that don’t show up in
camera receives light from the features of interest. visible light.

Introduction to Machine Vision 17

Image analysis Types of image analysis tools
Depending on the task, a vision system can leverage one or several specific image analysis tools.

software Edge detection/ Pixel counting

With the right combination of sensor, lens, lights,
extraction A straightforward algorithm
In an image, an edge can count the number of
and other components in place, a machine vision
marks a distinct change pixels at each grayscale
system is ready to capture images. But that’s only
in intensity. Such changes level in a region of interest
the first step.
mark discontinuities in depth, or the overall image,
Once it acquires an image, the system needs orientation, and material, providing information used
to process and analyze the result to make a and allow for finding the by other functions.
decision: to read a barcode, detect a defect, boundaries of objects.
confirm that a kit contains all of its items, or
measure a part.

Historically, the only way to get enough

processing power to do image analysis was to
Blob detection Pattern/template
send the image to an external PC. Depending
Different regions can matching
be identified, providing Guiding and positioning
on the size and complexity of the image, that
information used by other functions use pattern
could take time, potentially slowing down the line,
functions. matching or template
requiring a separate PC dedicated to the task.
matching on parts and
Today, highly capable vision systems have products to determine object
embedded processing, which speeds computation location and orientation.
and reduces possible points of failure.

Color analysis
Where object color is an
important feature, there
are several color tools,
including color extraction
and segmentation, and
color matching.

Introduction to Machine Vision 18

Putting image information to use
The information derived from that image analysis isn’t much good if it stays in the camera. It needs to be shared with other systems that can put it to use.

To do this, machine vision systems use standard industrial formats or protocols to share data with other devices, usually via wired connections. That data can then be used in many ways:

It can tell a nearby programmable logic controller (PLC) to fire a piston to bump It can be used with the factory’s process control system or a manufacturing
defective parts off the line. execution system (MES), so the system can make adjustments to optimize the
manufacturing process.

It can pop up on a human-machine interface (HMI) display so an operator can check It can trigger sorting mechanisms in an automated sortation system to get a package on
up on a production process and make adjustments. to the right truck.

Introduction to Machine Vision 19

Image-based data drives larger decisions, too Types of communication
In addition to individual actions on single parts, products, or packages, the data
generated through machine vision applications can also be aggregated and used to protocols
tackle larger business issues. There is a wide range of protocols used in
automated manufacturing, from simple on/
For example, businesses can use the data to find the root causes behind delays,
off signals sent by a direct connection, to
defects, and missed shipments. Is one supplier using inferior raw materials for their part?
sophisticated industrial protocols such as
Or is the problem caused by how parts are stored or shipped? Are workers being trained
Ethernet/IP, Profibus, and DeviceNet, among
inadequately? Machine vision data can be analyzed to identify the source of the problem.
many others.
Other uses of machine vision data can include:
While interoperability is common, meaning you
can buy the best device for a specific job and
be confident it can communicate with all your
current equipment, it remains true that various
manufacturers favor one protocol or another,
and there are a lot of proprietary standards.

Machine vision systems can be configured

to be interoperable with virtually any set of
industrial protocols, providing their information
in a form that is easily used to improve
automated industrial operations.
Learning that a certain type of part causes rejection of Locating and identifying parts, constantly updating
the assemblies it is installed in. inventory, and guaranteeing accurate order fulfillment.

Analyzing large data sets with an enterprise resource Reviewing failure rates and types to pin down where
planning (ERP) system to make long-term operational productions errors are being introduced.
decisions.

Introduction to Machine Vision 20

Conclusion
Industrial automation enabled by machine vision is
increasingly essential for competing in the market and
meeting compliance requirements. Machine vision
reduces costs, generates data for process improvement,
increases OEE, enables new types of automation, and
improves safety.

Machine vision constantly improves, both through

technical innovations such as higher-resolution sensors,
improved lenses, more capable processors, and smaller,
more rugged casings, as well as software advances
such as new AI algorithms that make machine vision
more accurate, faster, and easier to use.

As these advances continue, machine vision systems

will be used in more ways, by more types of businesses,
with greater impact on overall results.

Introduction to Machine Vision 21

 www.cognex.com

Companies around the world rely on Cognex vision and barcode reading
solutions to optimize quality, drive down costs, and control traceability.

Corporate Headquarters One Vision Drive Natick, MA 01760 USA

Regional Sales Offices Ireland +353 21 421 7500 India +91 7305 040397
Italy +39 02 3057 8196 Indonesia +62 21 3076 1792
Americas Netherlands +31 207 941 398 Japan +81 3 5977 5400
North America +1 844 999 2469 Poland +48 717 121 086 Korea +82 2 539 9047
Brazil +55 11 4210 3919 Romania +40 741 041 272 Malaysia +6019 916 5532
Mexico +800 733 4116 Spain +34 93 299 28 14 New Zealand +64 9 802 0555
Europe Sweden +46 21 14 55 88 Phillipines +63 2 8539 3990
Austria +43 800 28 16 32 Switzerland +41 445 788 877 Singapore +65 3158 3322
Turkey +90 216 900 1696 Taiwan +886 02 7703 2848 © Copyright 2024, Cognex Corporation. All information
Belgium +32 289 370 75
United Kingdom +44 121 29 65 163 Thailand +66 6 3230 9998 in this document is subject to change without notice. All
Czechia +420 800 023 519
Vietnam +84 98 2405167 Rights Reserved. Cognex is a registered trademarks of
France +33 1 76 54 93 18 Asia-Pacific Cognex Corporation. All other trademarks are property of
Germany +49 721 958 8052 Australia +61 2 7202 6910 their respective owners. Lit. No. ItMV-EN-02-2024
Hungary +36 800 80291 China +86 21 5875 1133 www.cognex.com