September 2022 - 10

feature | digital transformation
can analyze trends, using this feedback to
make corrections to the process parameters
online-with minimal or no operator or
quality personnel intervention.
TECHNOLOGIES USED
An MV system's primary component
is a vision sensor or industrial camera.
Depending on the application,
sensors capable of operating within
specifi c electromagnetic spectrum
bands can be chosen to best show the
features of interest with maximum contrast.
Most applications implement sensors operating
in the visible light range of 380 to 700
nanometers, like the human eye. Applications
that require inspecting the integrity
of a thermal signature implement a camera
operating in the medium-wave infrared
(MWIR) range. Examples include inductively
sealed caps on food/beverage and
pharmaceutical packaging or the heating
coil for a car seat. Pharmaceutical, produce,
and meat inspection applications may implement
hyperspectral imaging technologies
to isolate the defects with optimal accuracy.
Sensors typically acquire images in a rectangular
format. Th ese are called area-scan
images; the sensors are categorized as
area-scan sensors. Specialized applications
such as continuous-web, round-bottle, or
label inspections use a specialized line-scan
sensor. Th e images are generated by combining
individually triggered, one-dimensional
line images. Area-scan and line-scan
sensors generate 2D images.
Advanced applications use 3D sensors
to generate a point-cloud representation of
the object. Th e point-cloud or the heightto-grayscale
representation can be used to
process 3D images and measure features of
varying heights. Typical applications for this
technology include pallet inspection, package
dimensioning, feature-height measurements
for automotive parts, tire DOT code
and tread inspection, surface roughness
10 | EFFICIENTPLANTMAG.COM
false-failure rates for MV systems to
minimize waste. DL deployments typically
require a large, good-quality dataset to train
accurate production-deployment models.
Such datasets are meticulously and
manually generated by personnel who
know the defect representations and
the industrial process. Th is can become
cumbersome and result in project time
and capital overruns. Deployment of
such algorithms typically takes longer
Advanced applications use 3D sensors to generate
a point-cloud representation of the object
such as this engine. Image: Motion Ai
inspection, and robot guidance.
Rules-based computer vision algorithms
are extremely powerful and solve most
applications. Th ey typically need a small
set of images and image-processing servers
without GPUs to deploy in production.
However, these solutions may become
highly complex and hard to maintain and
scale when production shows variation. One
example is OCR applications; training and
maintaining them is prohibitively complex
when additional fonts are introduced. Such
scenarios have a high likelihood of reduced
accuracy, a huge risk for validated systems,
especially in the pharmaceutical, automotive
and aerospace industries.
DEEP-LEARNING APPLICATION
Th e application of DL models has proven
eff ective due to the limitations of rulesbased
computer-vision algorithms, plus
the increasing demand for solving some of
the tougher MV applications, specifi cally
classifi cation and anomaly detection. Deep
learning falls into the machine-learning
fi eld, all under the scientifi c discipline of AI.
Industries require extremely low
to validate due to their iterative nature.
In the current state, deep learning implementations
are best suited to solve narrow
problems, which are extremely diffi cult for
rules-based algorithms. Hybrid algorithms
implementing rules-based algorithms and
DL models-leveraging the strengths of
both-have proven immensely successful.
Machine-vision systems can play a highly
eff ective role in making modern manufacturing
more effi cient. Th ese systems can
be implemented at almost every stage of
manufacturing across various industries and
serve as hubs that generate rich data to help
quality engineers, manufacturing engineers,
production managers, and executives better
understand the process. MV systems are
indispensable in the modern factory, eliminating
slow and inaccurate manual inspection
processes. AI and DL technologies add
capabilities to tackle some of the toughest
inspection requirements while maintaining
or improving existing accuracy. EP
Navneet Nagi is a Senior Machine Vision
Application Engineer at Integro Technologies
Corp., a division of Motion Ai, Birmingham,
AL (ai.motion.com). He has developed,
scaled, and helped deploy numerous
automated inspection systems across industries
ranging from agriculture to aerospace.
Nagi has extensive experience developing algorithms
for automated industrial inspection
applications using computer vision techniques
and deep learning.
SEPTEMBER 2022
http://ai.motion.com http://www.EFFICIENTPLANTMAG.COM

September 2022

Table of Contents for the Digital Edition of September 2022

September 2022 - Cover1
September 2022 - Cover2
September 2022 - 1
September 2022 - 2
September 2022 - 3
September 2022 - 4
September 2022 - 5
September 2022 - 6
September 2022 - 7
September 2022 - 8
September 2022 - 9
September 2022 - 10
September 2022 - 11
September 2022 - 12
September 2022 - 13
September 2022 - 14
September 2022 - 15
September 2022 - 18
September 2022 - 17
September 2022 - 18
September 2022 - 19
September 2022 - 20
September 2022 - 21
September 2022 - 22
September 2022 - 23
September 2022 - 24
September 2022 - 25
September 2022 - 26
September 2022 - 27
September 2022 - 28
September 2022 - 29
September 2022 - 30
September 2022 - 31
September 2022 - 32
September 2022 - Cover3
September 2022 - Cover4
https://www.nxtbookmedia.com