October 17
27
E
mbEddEd
C
omputing
Omron OKAO vision software, a proven set
of image recognition algorithms used in over
500 million digital cameras, mobile phones
and surveillance robots around the world. Key
features of the module include speed and
consistency of response, and the distance
over which it can take readings. For exam-
ple, HVC can capture, detect and recognise a
face and provide information like age, gender
and mood over a distance of 1.3m in 1.1s and
will provide a confidence level with its read-
ing. HVC implements the OKAO software on
a hardware platform complete with camera,
processor and data interface optimised specif-
ically in terms of its digital and optical design
for this application. The module is a fully
integrated, plug-in solution. The developer
can just look at the outputs and configure the
system to make appropriate decisions
depending on their status. Intelligent vending
machines offer an even wider range of goods,
and need to adapt quickly to changing retail
needs. Drinks vending machines offer a par-
ticular challenge, as they need to detect the
presence and level of different liquids, some
of which may be entirely clear, and the pres-
ence or absence of cups, which again could be
clear plastic. The fact that these liquids may
be hot and pose a safety hazard increases the
importance machine manufacturers attach to
accurate and reliable detection.
Although conventional photo micro sensors
(PMS) are often suitable, it is well known
that they struggle with many types of surface.
Transparent objects and liquids have always
been hard to detect, as have mirrored, dif-
fuse and deep black surfaces. The new Omron
B5W Light Convergent Reflective Sensor
(figure 2) detects target objects much more
reliably than general reflection photo sensors,
even those that are often a challenge such as
reflective, transparent, diffuse or black sur-
faces. The sensor convergent light beam can
be accurately set to trigger on the presence of
an object within a tightly specified target area,
and ignore any objects in the background or
foreground. These sensors are equally suitable
for contactless detection of clear or coloured
liquids in transparent tanks. The B5W fea-
tures a combination of a cylindrical and a
non-spherical lens.
Despite the increasing popularity of contact-
less payment, many customers still like to
present cash. Inflation means that increas-
ingly, machines need to accept notes and to
return them in change. The latest high-reso-
lution micro displacement sensors can greatly
increase the reliability of these mechanisms,
offering contactless measurement of the
thickness of paper and multi-feed detection
through minute changes in the location of
light reflected from the object.
For example, the new Z4D micro-displace-
ment sensor (figure 3) from Omron is capa-
ble of resolving just 10 microns – one tenth
the diameter of a human hair or the thickness
of a coat of paint. The exceptional resolution
is based on a proprietary optical design using
the triangulation. This structure achieves very
good and stable linearity between the output
voltage and distance, simplifying the design
of the host circuit. Like other electronic sys-
tems, intelligent vending machines are subject
to power control directives, like the EU ErP
Ecodesign Directive which specifies that the
power consumption of equipment in standby
mode should not exceed 0.50W, unless the
equipment has a status display in which case
1W is permissible. This issue applies partic-
ularly to tabletop vending machines such as
coffee machines issued to offices, which are
used less intensively and can be left to switch
off when not in use, overnight for example.
To support this requirement, Omron has
introduced the industry-smallest remote-re-
set rocker switch supporting zero-standby
power design. This switch can be controlled
remotely and is offered in a version with a
delayed-off feature to support safe system
shut-down. With this switch, the power to
the system is maintained when the switch is
returned to the off position. The system is
then powered down by an external signal after
an interval determined by the designer. This
prevents issues with data loss or circuit dam-
age through forced power-off by the user, and
can help safer application design.
Figure 2. The B5W light convergent reflective sensor detects target objects even those that are often
a challenge such as reflective, transparent, diffuse or black surfaces.
Figure 3. The Z4D micro-displacement sensor is capable of resolving just 10 microns.
