White Paper
Industry Report:
Sorting Fresh-Cut Produce
Industry Report: Sorting Fresh-Cut Produce
Until recently, processors of fresh-cut produce have relied on labor-intensive
manual inspection to remove defects and foreign material (FM). But tightening
restrictions on pesticide use and the growth of organic products are making
defects more common while the scarcity and cost of labor and consumers’
increasing scrutiny of fresh-cut product quality is rising. Given these market
dynamics, processors are looking for methods to improve inspection.
Advancing technology and experience in other food segments have ushered
in a new set of solutions. Automated optical inspection systems (also called
sorters), which have been widely adopted for decades in the potato processing
and processed vegetable industries, have recently been developed for fresh-cut
produce. Compared to manual inspection, which is inconsistent and subjective,
sorters are able to assure product quality and food safety by more effectively
identifying and removing defects and foreign material, while at the same time
reducing labor costs and improving operating efficiencies.
In this white paper, we will explore a wide range of sorting technology. The
objective is to help fresh-cut processors understand what tools can be leveraged
to maximize product quality and identify the criteria they should consider when
selecting the ideal sorter for their products and applications.
Sorting Basics
From on-belt to in-air systems, sorters typically
handle up to 7.5 metric tons per hour when sorting
leafy greens and up to 28 metric tons per hour
when sorting cut vegetables. Some sorters rely
on cameras, others on lasers, and some combine
cameras and lasers to view product from the top
only or both top and bottom. Some sorters inspect
only an object’s color, others inspect an object’s
color, size, and shape, and some sort based on the
object’s structural properties, including differing
levels of chlorophyll. The processor’s products and
business objectives determine the suitable sorter
configuration.
Regardless of configuration, each sorter contains
similar basic elements. The material handling
component presents the product in the optimal
© Key Technology, Inc.
2
Industry Report: Sorting Fresh-Cut Produce
method to the sensors. The sensors capture data, which is analyzed by the imageprocessing system. Defective product and foreign material are either ejected by
mechanical paddles or air jets.
Although sorters are designed for continuous, 100 percent, in-line inspection at full
production speeds, they can also be used in a batch-feed mode.
Cameras and Lasers and Wavelengths
The ideal sorter for any given application combines the lights, cameras, lasers, and
image processing software that most effectively differentiate good product from defects
and foreign material. To maximize that differentiation, it is important to identify the
wavelengths that produce unique “signatures” for each object of interest. The sorter
manufacturer might use a spectrophotometer on the customer’s products, defects and
foreign material to see how these objects respond to different wavelengths. Armed with
this data, the manufacturer will identify the ideal wavelengths or sets of wavelengths
from infrared (IR) to visible to ultraviolet (UV) for the specific application and recommend
the most appropriate technology to achieve the desired results.
Tri-chromatic cameras can be set to inspect within the visible range (red, green, and
blue) or a combination of visible and IR or UV spectrums. These cameras capture product
information based primarily on material reflectance and, depending on the image
processing software, can recognize defects and foreign material based on color, size, and
shape.
Sorting systems can also use lasers, sometimes in conjunction with cameras, to inspect
product. Lasers are used primarily to inspect a material’s structural properties, which make
them ideal for detecting a wide range of foreign material and some product defects. Like
© Key Technology, Inc.
3
Industry Report: Sorting Fresh-Cut Produce
cameras, lasers can be designed to
inspect only within the visible range
or within the IR or UV spectrums too.
If multiple sets of wavelengths are
needed to detect the range of defects
and foreign material for any given
product, the sorter can be equipped
with multiple cameras set at differing
wavelengths and/or multiple lasers
or a combo-laser that simultaneously
inspects at more than one set of
wavelengths. Similarly, if multiple
products are run on the line and each product is most effectively inspected at a unique
wavelength, selecting a sorter that inspects at all these wavelengths is desirable.
Size, Shape, and Color
All sorters, even the simplest systems that rely only on monochromic (black and white)
cameras, can detect differences in color (if only on the gray scale) to distinguish good
product from defects and foreign material. But most sorters are capable of much more.
Sophisticated color cameras are capable of detecting millions of subtle color differences
to better distinguish good from bad objects. And the resolution of cameras and lasers
differ with the highest resolution sensors able to detect the smallest defects and foreign
material. The resolution of commonly available cameras and lasers detect defects and
objects down to 3-5mm. Ultra-high resolution systems can detect defects and objects as
small as 1mm.
Enhanced capabilities can be added if the sorter’s image processing software offers
“object-based recognition,” enabling it to analyze objects based on size and shape as
well as the location of the defect on the product, if desired. Some sorters even allow the
© Key Technology, Inc.
4
Industry Report: Sorting Fresh-Cut Produce
user to define a defective product based on the total defective surface area of any given
object. These object-based considerations put more power into the processor’s hands to
produce optimal product quality.
Fresh-Cut Applications
A wide range of leaf defects can be
identified by sorters equipped with
color cameras. If the defects can be
seen on both the sides of the product, a
sorter with only top-mounted cameras
is effective. To detect and remove
single-sided defects, and in situations
where product overlap occurs at
higher capacities, sorters with top- and
bottom-mounted cameras are often
recommended.
Defects associated with water
exposure, sun exposure, chemical burn,
insect damage, rodent damage, rot,
disease, bacteria, and fungus, as well as problems in the outer wrap of iceberg, romaine,
and cabbage due to bruise damage or wilt, can all be removed with color camera-based
sorters. Typically, color cameras that inspect within the visible spectrum are most effective
for detecting leaf defects in iceberg, romaine, and cabbage. Vis/IR (a combination of
visible and IR) cameras are usually most effective for baby spinach and spring mix.
But much more is possible with color sorting. One processor that packs peach slices in
glass jars learned that customers prefer the color of the slices to be consistent. Mix yellow
and orange slices in one jar and customers perceived the yellow slices as unripe and left
the jar on the shelf. This processor used color sorting to separate the slices by color. The
technology allowed them to pack jars with only yellow slices and jars with only orange
slices. All the jars sold well and their sales increased.
Shape sorting has been used in the processed vegetable industry for years to differentiate
green beans from same-color stems and knuckles. Extend this shape-sorting capability
further and consider using the technology to separate straight green beans from curved
ones. Such a separation would enable the processor to package straight beans in single
serve packs and price them at a high mark-up while diverting curved beans to bulk
product, thus increasing the overall value of the green beans.
Processors of leafy greens such as iceberg, romaine, cabbage, spinach, spring mix, mâche,
butter leaf, arugula, and oakleaf often find sorting with a combination of cameras and
lasers most effective. The cameras detect leaf defects based on color while the lasers
detect insects and animal parts as well as sticks, rocks, cardboard, plastic, metal, and glass,
even if they are the same color as the good product, based on the object’s structural
properties.
© Key Technology, Inc.
5
Industry Report: Sorting Fresh-Cut Produce
One type of laser sorter that is very effective for a variety of fresh-cut products is
fluorescence-sensing laser sorters. These sorters detect objects’ differing levels of
chlorophyll to detect and remove foreign material. This technology is so powerful, it can
sort leafy greens and identify and remove leafy green product left over from the prior
crop as well as leaves from trees, even if they have similar color, texture, and shape as
good product.
Fluorescence-sensing
laser sorters are
also useful to some
processors of cut
vegetables. For
example, carrot
processors interested
in identifying and
removing carrot tops
with stems remaining
on the crown or
even embedded
in the crown can
achieve this sort with
fluorescence-sensing
lasers.
Sorters that combine
color cameras and
infrared lasers can be
effectively used to
detect and remove
core from cut iceberg
and romaine lettuce,
along with leaf defects and foreign material. This powerful capability allows processors
to cut the un-cored head with conventional cutting technology and then use the sorter
to remove the core. It enables processors to eliminate manual cutting and outer leaf
removal, which reduces labor costs and improves yields. And because the core is not
removed in the field, the shelf life of the product improves.
Sorter Selection Criteria
When searching for the perfect sorter for any given application, several variables should
be considered beyond throughput, cameras, lasers, and wavelengths.
The value of the sorter manufacturer’s experience cannot be underestimated. Their
expertise helps identify the ideal wavelengths and sensors to use to achieve the customer’s
sorting objectives given the products and applications. Their expertise also guides them
to consider custom-engineered product handling components that minimize product
damage and sanitation features such as clean-in-place systems that minimize bacteria and
© Key Technology, Inc.
6
Industry Report: Sorting Fresh-Cut Produce
keep the sorter operating at peak performance.
Of course, the effectiveness of the sorter relies not only on the hardware but on the
software – the algorithms – that manipulate raw data and categorize information based
on the customer-defined accept/reject thresholds. The art and science of image processing
lies in developing computerized routines that improve the effectiveness of the operation
while presenting a simple user-interface to the operator. Thus, the sorter manufacturer’s
expertise in developing algorithms for the customer’s products affects both the sorter’s
performance and ease of use.
When comparing competitive systems, consider the resolution of the cameras and
lasers because higher resolution allows the sorter to detect and remove smaller defects.
Compare cameras and their ability to detect possibly millions of subtle color differences.
Compare the illumination system (usually either fluorescent, LEDs, or HID), understanding
that superior lighting leads to superior sorter performance.
Sorters are sophisticated pieces of equipment based on technology that advances at a
rapid rate. As technology advances, the capabilities of sorters grow, which can be used to
the processor’s advantage. To continue to get the most from a sorter and maximize the
return on investment, look for a modular sorter that is designed to be easily upgraded or
reconfigured in the field.
Last but not least, it is important to consider the level of service a supplier can provide in a
specific region – from engineering to after-sales support.
The Bottom Line
With the arrival of sorters designed specifically for
fresh-cut produce, processors now have a highly
effective tool for removing defects and foreign
material while reducing labor costs and improving
operating efficiencies. Processors that select and
install the ideal sorter for their application are
better able to consistently assure product quality
and food safety. But most importantly, they are
safeguarding their customers and protecting their
brands.
Published by:
Key Technology, Inc.
T
509.529.2161
150 Avery Street
E
[email protected]
Walla Walla, WA 99362
www.key.net
© Key Technology, Inc.
7