- Library Home /
- Search Collections /
- Open Collections /
- Browse Collections /
- UBC Undergraduate Research /
- Real-Time Image Processing in Mechatronics Systems...
Open Collections
UBC Undergraduate Research
Real-Time Image Processing in Mechatronics Systems – (RIMS) Keshmiri, Saman; Issaei, Amir; Willems, Robert
Abstract
Rodents have become a problem in a older greenhouse at the UBC Farm. They find the warm atmosphere of the greenhouse and the harvest and seedlings stored in the greenhouse a welcoming home. The rodents damage the hard work of the UBC Farm staff. Conventional rodent control mechanism have proven ineffective such as sonic traps which use high frequency ultrasound; snap traps require a great deal of attention and attract birds into the greenhouse when rodents are caught; resealing the greenhouse is far too costly; thus, the our team has designed an active device to track rodent movement within the bounds of the greenhouse and follow the rodents with the ability to trigger an external device for frightening the rodents. The external devices is out of the scope of our engineering team’s knowledge and is left up to professionals in the field to implement. We present a real-‐time image-‐processing mechatronic system. Using an infrared camera, to eliminate the distinction between daylight and night time, to capture images of the greenhouse and using image processing on the computer track rodents. The device installed in the greenhouse includes a housing mechanism for the camera with motors to control the position of the camera and consequently any external triggering device attached to it. The tracking system is tested in the laboratory and has proven effective at tacking small objects in both regular light and low light conditions. There is a ~1.5s delay in motion present in front of the camera and reaction of the tracker; however, in the time frame present our team was not able to eliminate this delay completely but reduced it down from 2.8s to 1.5s using sophisticated probabilistic models for immediate future motion tracking. The system has 5 components, 3 of which are coded on a computer using MATLAB computation software while the other 2 components are an Arduino microcontroller and the infrared camera. The computer and microcontroller communicate via a serial connection while information from the camera is read into the computer via Ethernet. Accurate and precise motion detection allows the system to detect the smallest of changes in the environment it is monitoring thus making the system ideal for small rodents.
Item Metadata
| Title |
Real-Time Image Processing in Mechatronics Systems – (RIMS)
|
| Creator | |
| Date Issued |
2012-01-09
|
| Description |
Rodents
have
become
a
problem
in
a
older
greenhouse
at
the
UBC
Farm.
They
find
the
warm
atmosphere
of
the
greenhouse
and
the
harvest
and
seedlings
stored
in
the
greenhouse
a
welcoming
home.
The
rodents
damage
the
hard
work
of
the
UBC
Farm
staff.
Conventional
rodent
control
mechanism
have
proven
ineffective
such
as
sonic
traps
which
use
high
frequency
ultrasound;
snap
traps
require
a
great
deal
of
attention
and
attract
birds
into
the
greenhouse
when
rodents
are
caught;
resealing
the
greenhouse
is
far
too
costly;
thus,
the
our
team
has
designed
an
active
device
to
track
rodent
movement
within
the
bounds
of
the
greenhouse
and
follow
the
rodents
with
the
ability
to
trigger
an
external
device
for
frightening
the
rodents.
The
external
devices
is
out
of
the
scope
of
our
engineering
team’s
knowledge
and
is
left
up
to
professionals
in
the
field
to
implement.
We
present
a
real-‐time
image-‐processing
mechatronic
system.
Using
an
infrared
camera,
to
eliminate
the
distinction
between
daylight
and
night
time,
to
capture
images
of
the
greenhouse
and
using
image
processing
on
the
computer
track
rodents.
The
device
installed
in
the
greenhouse
includes
a
housing
mechanism
for
the
camera
with
motors
to
control
the
position
of
the
camera
and
consequently
any
external
triggering
device
attached
to
it.
The
tracking
system
is
tested
in
the
laboratory
and
has
proven
effective
at
tacking
small
objects
in
both
regular
light
and
low
light
conditions.
There
is
a
~1.5s
delay
in
motion
present
in
front
of
the
camera
and
reaction
of
the
tracker;
however,
in
the
time
frame
present
our
team
was
not
able
to
eliminate
this
delay
completely
but
reduced
it
down
from
2.8s
to
1.5s
using
sophisticated
probabilistic
models
for
immediate
future
motion
tracking.
The
system
has
5
components,
3
of
which
are
coded
on
a
computer
using
MATLAB
computation
software
while
the
other
2
components
are
an
Arduino
microcontroller
and
the
infrared
camera.
The
computer
and
microcontroller
communicate
via
a
serial
connection
while
information
from
the
camera
is
read
into
the
computer
via
Ethernet.
Accurate
and
precise
motion
detection
allows
the
system
to
detect
the
smallest
of
changes
in
the
environment
it
is
monitoring
thus
making
the
system
ideal
for
small
rodents.
|
| Genre | |
| Type | |
| Language |
eng
|
| Series | |
| Date Available |
2012-09-20
|
| Provider |
Vancouver : University of British Columbia Library
|
| Rights |
Attribution-NonCommercial-NoDerivatives 4.0 International
|
| DOI |
10.14288/1.0074479
|
| URI | |
| Affiliation | |
| Campus | |
| Peer Review Status |
Unreviewed
|
| Scholarly Level |
Undergraduate
|
| Rights URI | |
| Aggregated Source Repository |
DSpace
|
Item Media
Item Citations and Data
Rights
Attribution-NonCommercial-NoDerivatives 4.0 International