machine would not only take up additional
space, but would further increase costs by
requiring tube-shoulder sorting and singulation for inspection.
After preparing machine plans and cre-
ating custom bracket holes, four BFLY-
PGE-50A2C-CS cameras from FLIR Inte-
grated Imaging Solutions (formerly Point
Grey, Richmond, BC, Canada; www.ptgrey.
com) were mounted directly on a VIC TORY
180 molding machine from ENGEL (Schw-
ertberg, Austria; www.. engelglobal.com).
With a resolution of 2592 x 1944 pixels and a
frame rate of 13 frames per second, the cam-
eras are powered using PoE – Power Over
Ethernet and connect via 4x10Gbit connec-
tion to a powerful VISTION industrial PC
with a 3120ix Intel Core i7 ( 3. 3 GHz), 4 GB
of DDR3 ram and a 256 GB SSD solid state
drive, which controls the cameras and illu-
The biggest challenge with integrating
the solution on an existing injection molding
machine were proper
before it is har-
vested, thus improving yield. Automated
harvesting systems using machine vision
can perform this task. Using 2D vision sys-
tems that employ cameras mounted in front
of a tractor are, however, subject to ambi-
ent lighting conditions.
To develop algorithms to detect, identify
and size such products is complex. Typically,
Since 2D systems
are also calibrated
to a particular plane,
products of vari-
able height make 2D
image processing sys-
tems used for such
tasks more complex.
However, with a pre-
calibrated 3D imag-
ing system, the angle
and tilt of the prod-
uct can be analyzed
so that before the product is picked, a robot
can be fed the correct angular coordinates
such that the product will not be damaged.
Currently, such 3D automated harvest-
ing systems are being developed by Cap-
ture Automation (Hove, East Sussex, Eng-
land; www.captureautomation.co.uk). Using
a combination of 3D laser scanners, robotics,
image processing and deep learning soft-
ware the company is developing systems to
automate the harvesting of broccoli heads.
One of the advantages of 3D laser scanners is that they can be used continuously to
capture images that can be processed on-the-fly. Using an encoder mounted on the
tractor, positional coordinates can be sent
to the robot, track the product as it moves
under the robot head and then pick them.
To gain such accurate information from a
moving system, an accurate encoder must
be used. In the first prototype of the system,
a rotary encoder wheel was positioned at
the front of the tractor.
Unfortunately, in rainy weather, such
rotary encoder wheels may slip rendering
the calibration of the system ineffective. To
overcome this, a spike-based encoder was
developed that provides more accurate positional information.
The harvester can then be used to select
different types of broccoli heads according
to their weight and size and place each dif-
ferent head into a different bin. The system
can also report what heads may be left in
the ground or ones that may be suitable for
picking later. While
Data generated in the form of a height map from the Gocator is gen-
erated as a grey-scale image (left) where the lighter the pixel data the
closer the broccoli head is to the camera and vice-versa.
The VISTION d.o.o. company along with its client the SIBO GROUP have retrofitted an injection molding machine with a vision system that performs in-mold (image a on left) inspection to detect molded tube shoulder defects (image b on right).
continued on page 13
continued on page 15
continued from page 11