Title: Systems and Methods for Defect Sorting of Produce
Field of the invention '
[0001] The invention generally relates to ways of identifying undesirable produce, and, in particular to systems and methods for identifying and discarding undesirable pieces of produce. Background of the invention
[0002] In an agribusiness, a typical step in the process of sorting produce is to identify spoilt, damaged and/or otherwise grotesquely defective pieces of produce intermixed with acceptable pieces of produce. After the unacceptable items are identified they are discarded so as not to further contaminate the acceptable produce or be passed along to consumers. The sorting of produce is typically done in an assembly line fashion and manual inspection of the produce by people is still often used to identify and remove unacceptable items. In order to expedite this task, the bulk of the work has been automated through the use of computer-controlled equipment. [0003] Typical defect-sensing equipment includes cameras and/or photo-sensors that optically scan individual pieces of produce. The images captured by the cameras and/or photo-sensors are processed by software that, on a pixel level, attempts to identify defects that are characterized by relative discoloration over a portion of the surface of a particular piece of produce. The software permits a margin (e.g. 5-15%) of color differential to avoid a high number of false positives. The problem is that defects, such as splits or surface abrasions in some produce - where the flesh color is similar to the skin color - cannot be accurately identified by this equipment, since the difference in skin color and flesh color falls within the margin of color differential allowed by the software. Reducing the margin would greatly increase the number of false positives, since most types of produce naturally have some degree of color non-uniformity in the skin to begin with.
[0004] Additionally or alternatively, in some cases infrared sensors/cameras are used in the process of identifying defective pieces of
produce. In order for infrared sensors/cameras to be effective, the skin of the produce must be heated by as much as 150C. Heating the produce may prematurely advance ripening and may even start to cook the produce. Neither result is desirable.
Summary of the invention
[0005] According to an aspect of an embodiment of the invention there is provided a method for identifying defects in produce including: applying a food safe fluorescing dye to produce to be inspected, the food safe fluorescing dye serving to enhance differences between defects and acceptable natural features present on produce under ultraviolet (UV) light; and, inspecting dyed produce under a UV light to identify possibly unacceptable pieces of produce.
[0006] In some embodiments, the inspecting of dyed produce under UV light is done manually by people, whereas in other embodiments the inspecting of dyed produce under UV light is carried out by an automated means that includes a controller connectable to one of a UV sensing device and a camera suitable for capturing UV images.
[0007] In some embodiments a method also includes a step of sorting unacceptable produce from acceptable produce under the UV light.
[0008] In some embodiments a method also includes a step of arranging produce to be inspected into one or more lanes.
[0009] In some embodiments a method also includes a step of washing the produce before applying the food safe fluorescing dye. [0010] In some embodiments a method also includes a step of rinsing the produce after applying the food safe fluorescing dye.
[0011] In some embodiments a method also includes a step of enforcing an absorption delay after the food safe fluorescing dye has been applied.
[0012] In some embodiments a method also includes a step of enforcing an adsorption delay after the food safe fluorescing dye has been applied.
[0013] In some embodiments a method also includes a step of manually pre-sorting the produce.
[0014] In some embodiments a method also includes selecting and providing a food safe fluorescing dye that preferably absorbs into exposed pulpy flesh of the produce to be inspected rather than into the skin of the produce. [0015] In some embodiments a method also includes selecting and providing a food safe fluorescing dye that preferably absorbs into exposed pulpy flesh of the produce to be inspected rather than onto the skin of the produce.
[0016] In some embodiments a method also includes the food safe fluorescing dye is expected to adhere to the flesh and the skin of the produce to various respective degrees depending upon the defects present and the acceptable natural features of the produce. Similarly, in other embodiments, the food safe fluorescing dye is expected to absorb into the flesh and adsorb onto the skin of the produce to various respective degrees depending upon the defects present and the acceptable natural features of the produce.
[0017] According to another aspect of an embodiment of the invention there is provided a system for identifying defects in produce including: an applicator for applying a food safe fluorescing dye to produce, wherein the food safe fluorescing dye is absorbed into produce to various degrees depending upon defects present and acceptable natural features of the produce and serves to enhance differences between defects and the acceptable natural features of the produce under ultraviolet (UV) light; and a UV light source for emitting at least UV light and thereby activating fluorescence from the food safe fluorescing dye.
[0018] In some embodiments a system also includes a UV light sensor for detecting fluorescence emitted from the food safe fluorescing dye.
[0019] In some embodiments a system also includes a controller, connectable to receive an input from the UV light sensor to determine whether or not the fluorescence emitted from the fluorescing dye is indicative of a defect or an acceptable natural feature of the produce being inspected.
[0020] In some embodiments a system also includes a computer readable program code means for identifying defects in produce, the computer readable program code means includes: instructions for optically scanning produce under UV light; and, instructions for parsing scanned images of the produce under the UV light to determine whether or not the fluorescence emitted from the fluorescing dye is indicative of a defect or an acceptable natural feature of the scanned produce. In some specific embodiments, the computer readable program code means further comprises instructions for indicating that a piece of produce is defective when it has been determined that the fluorescence emitted from the fluorescing dye is indicative of a defect in the produce. In some other specific embodiments, the computer readable program code means includes instructions for enforcing an absorption delay after the food safe fluorescing dye has been applied to the produce.
[0021] In some embodiments a system also includes a rinsing station after the applicator, for rinsing off excess dye. In some related embodiments, the rinsing station is located a delay distance from the applicator to allow the food safe fluorescing dye time to at least one of absorb and adsorb. [0022] According to another aspect of an embodiment of the invention there is provided a method of using a food safe fluorescing dye for identifying defects in produce, wherein the food safe fluorescing dye is absorbed into produce to various degrees depending upon defects present and acceptable natural features of the produce and serves to enhance differences between defects and acceptable natural features present on produce under ultraviolet (UV) light, including: applying a food safe fluorescing dye to produce to be
inspected; and inspecting dyed produce under a UV light to identify unacceptable pieces of produce.
[0023] In some embodiments a method also includes a step of sorting out unacceptable produce from acceptable produce under the UV light. [0024] In some embodiments a method also includes a step of arranging produce to be inspected into one or more lanes.
[0025] In some embodiments a method also includes a step of washing the produce before applying the food safe fluorescing dye.
[0026] In some embodiments a method also includes a step of rinsing the produce after applying the food safe fluorescing dye.
[0027] In some embodiments a method also includes a step of enforcing an absorption delay after the food safe fluorescing dye has been applied.
[0028] In some embodiments a method also includes a step of manually pre-sorting the produce.
[0029] According to another aspect of an embodiment of the invention there is provided a kit of parts suitable for upgrading a produce sorting system to identify produce defects using a combination of a food safe fluorescing dye and ultraviolet (UV) light, including: an applicator for applying a food safe fluorescing dye to pieces of produce; and a UV light source for emitting UV light and thereby activating fluorescence from the food safe fluorescing dye.
[0030] In some embodiments a kit of parts also includes a UV light sensor for detecting fluorescence emitted from the food safe fluorescing dye. In related embodiments a kit of parts also includes a controller, connectable to receive an input from the UV light sensor to determine whether or not the fluorescence emitted from the fluorescing dye is indicative of a defect or an acceptable natural feature of the produce being inspected. In other related embodiments, a kit of parts also includes a computer readable program code means for identifying defects in produce, the computer readable program
code means having: instructions for optically scanning produce under UV light; and instructions for parsing scanned images of the produce under the UV light to determine whether or not the fluorescence emitted from the fluorescing dye is indicative of a defect or an acceptable natural feature of the produce being inspected. In such embodiments, the computer readable program code means may also optionally include instructions for indicating that a piece of produce is defective when it has been determined that the fluorescence emitted from the fluorescing dye is indicative of a defect in the produce. [0031] Other aspects and features of the present invention will become apparent, to those ordinarily skilled in the art, upon review of the following description of the specific embodiments of the invention.
Brief description of the drawings
[0032] For a better understanding of the present invention, and to show more clearly how it may be carried into effect, reference will now be made, by way of example, to the accompanying drawings, which illustrate aspects of embodiments of the present invention and in which:
[0033] Figure 1 is a flow chart depicting the general steps provided in a first method of defect identification according to an embodiment of the invention;
[0034] Figure 2 is a flow chart depicting the general steps provided in a second method of defect identification according to an embodiment of the invention;
[0035] Figure 3 is a flow chart depicting the general steps provided in a third method of defect identification according to an embodiment of the invention;
[0036] Figure 4 is a flow chart depicting the general steps provided in a fourth method of defect identification according to an embodiment of the invention;
[0037] Figure 5 is a schematic diagram of a first system for defect identification according an embodiment of the invention;
[0038] Figure 6 is a schematic diagram of a second system for defect identification according to another embodiment of the invention; [0039] Figure 7 is a schematic diagram of a third system for defect identification according to another embodiment of the invention; and
[0040] Figure 8 is a schematic diagram of a fourth system for defect identification according to another embodiment of the invention.
Detailed description of the invention
[0041] The process of sorting out unacceptable produce from acceptable produce can be time consuming and costly. Attempts to automate this process are limited by the non-uniform nature of produce. That is, natural differences in color, size, and weight of a particular type of fruit, vegetable, legume, or fungus make it difficult to perform an accurate and cost effective quality assessment in an automated manner. Some embodiments of the invention provide a system and/or method that may improve the accuracy of defect identification in produce in an expedited manner. Specific embodiments include a process for enhancing differences between defects and natural acceptable features on a particular type of produce, which is facilitated by the use of a food safe dye that emits fluorescence under ultraviolet (UV) light. Preferably, the differences are enhanced so much that when examined under UV light, they are more easily identified than previously possible.
[0042] Referring to Figure 1 , shown is a flow chart depicting the general steps provided in a first method of defect identification according to an embodiment of the invention. Briefly, a fluorescing dye is applied to produce and then inspected under UV light. For some types of produce with waxy skin and pulpy flesh - such as avocados, tomatoes, apples and cherries - the fluorescing dye is absorbed by the pulpy flesh exposed by abrasions and splits and is not absorbed by the waxy skin. In some other types of produce -
such as potatoes, peaches and kiwis - the fluorescing dye is absorbed into or coats (i.e. adsorbs) the skin and the flesh to different degrees, but still effectively enhances the differences between defects and natural acceptable features (e.g. such as potato eyes, and dimples on peaches) when the produce is placed under UV light.
[0043] Specifically, starting at step 1-1 a fluorescing dye is applied to the produce to be inspected. At step 1-2, a UV light is shone on the produce to activate the fluorescence of the fluorescing dye thereby enhancing the difference between defects and natural acceptable features on a particular type of produce. Subsequently, at step 1-3 unacceptable produce can more easily be separated from the acceptable produce, since defects are easier to identify.
[0044] Steps 1-1 to 1-3 can be carried out manually or in an automated manner employing equipment configured for the task or by a suitable combination of manual effort and automated means. As will be described further below, in some embodiments, an automated system includes a controller in managerial communication with a number of devices, including for example, sprayers, conveyor belts, rollers, UV lights and sensors.
[0045] Referring now to Figure 2, shown is a flow chart depicting the general steps provided in a second method of defect identification according to an embodiment of the invention. The steps included in the second method include all of the steps of the first method described above. Steps 2-1 , 2-3 and
2-4 correspond to steps 1-1 , 1-2 and 1-3, respectively. The difference provided in the second method is that at a step 2-2, between steps 2-1 and 2- 3, the produce undergoes a sub-process of singulating, sometimes referred to as singulation. The sub-process of singulation results in the arrangement of the produce into one or more lanes. Individual pieces of produce are distinctly separated from neighboring pieces of produce in the same lane, and in some specific instances neighboring pieces of produce are separated by near-equal distances. This sub-process is typically carried out on specially shaped roller beds and/or a conveyor system. Equivalent^, singulation may also be carried
out manually for delicate produce (e.g. grapes to be used with estate wines), although this would be rather time consuming. Singulation permits ease of inspection whether it be done using computer controlled equipment or manually. In alternative embodiments, the singulation sub-process may be done before and/or simultaneously along with the application of the fluorescing dye.
[0046] Referring now to Figure 3, shown is a flow chart depicting the general steps provided in a third method of defect identification according to an embodiment of the invention. The steps included in the third method include all of the steps of the first method described above. Steps 3-1 , 3-4 and 3-5 correspond to steps 1-1, 1-2 and 1-3, respectively. Additionally, the third method includes steps 3-2 and 3-3 between steps 3-1 and 3-4. At step 3-2, after the fluorescing dye has been applied, an absorption delay of a predetermined duration is enforced, thereby allowing the fluorescing dye to be absorbed. Subsequently, at step 3-3 excess fluorescing dye is rinsed away and the produce is dried to the extent of removing relatively large droplets of liquid. Steps 3-2 and 3-3 are not essentially done in combination and one step may be done without the other in various alternative embodiments. Additionally, in some alternative embodiments singulation of the produce may also be performed in combination with the steps of the third method.
[0047] As one specific example, provided herein is a brief description of how the third method may be used to identify and sort out damaged and/or rotting strawberries from strawberries of acceptable quality. In this example, the fluorescing dye is quinine carried in tonic water (the soft-drink commonly mixed with gin). The tonic water is sprayed on the strawberries (which have a somewhat waxy skin) and is allowed time to absorb into abrasions, splits, cracks and the like if they exist. If such damage does not exist the tonic water carrying quinine will form droplets on the surface of the strawberries, leaving coalesced spots of quinine on the skin of the strawberries. If the strawberries are not rinsed, the quinine spots on the skin of the strawberries are indistinguishable from quinine absorbed into damaged areas. Accordingly, the
strawberries are rinsed with clean water and dried (to remove relatively large droplets, which may take a few seconds or longer) so that only the quinine absorbed into damaged areas is seen under the UV light.
[0048] In a related alternative example, the fluorescing dye is quinine sulphate that is mixed with water at 100 parts per million. In this embodiment, rinsing is not needed, as this formulation does not seem to form large enough droplets; and, any residual left on the skin of the strawberries (from this specific formulation) does not significantly affect the accuracy with which unacceptable pieces can be identified. [0049] Referring now to Figure 4, shown is a flow chart depicting the general steps provided in a fourth method of defect identification according to an embodiment of the invention. The steps included in the fourth method include all of the steps of the first method described above. Steps 4-3, 4-5 and 4-6 correspond to steps 1-1 , 1-2 and 1-3, respectively. [0050] Starting at step 4-1 , produce is manually pre-sorted. Pre-sorting can be done manually by people in an assembly line manner or by machines that roughly sort through the produce. Pre-sorting allows debris and obviously unacceptable produce to be sorted from the produce and, in some instances also serves as the beginning of a singulation sub-process. After the pre- sorting, the remaining produce is washed and dried at step 4-2 before the fluorescing dye is applied at step 4-3.
[0051] Similar to the third method, the fourth method includes an absorption delay, at step 4-4, of a predetermined duration that is enforced, thereby allowing the fluorescing dye to be absorbed and/or adsorbed. Subsequently, at step 4-5 the produce is illuminated with UV light and defective produce is identified and sorted out at step 4-6. After the defect sorting, the remaining produce is further sorted and packaged, according to other characteristics that include, for example, weight, size and color.
[0052] The four methods described above are only very specific examples according to embodiments of the invention. Those skilled in the art
will appreciate that various other method steps may be added in combination with the method steps described and/or the method steps described may be combined in other combinations not including additional steps. For example, other method steps may be added such as optional rinsing and/or drying using blowers, sponge rollers, or other drying schemes available in the art. Alternatively, produce may be rinsed but not dried or produce may be assumed to be wet and dried without rinsing, etc. Moreover, the UV light may originate from a source that emits bands of light other than UV light in addition to UV light. [0053] The application of the food safe fluorescing dye may be done in a number of different ways. For example, produce may be: submersed in a tank containing the dye; sprayed with the dye as it traverses a conveyor or roller system; and/or, passed through an overhead wash. Additionally and/or alternatively, the dye in a vapor form may be condensed onto the produce. [0054] There are a number of food safe fluorescing dyes suitable for this process. For example, as noted above, quinine carried in tonic water and quinine sulphate diluted in water are suitable, as they emit a blue-white fluorescence under UV light. Additionally, turmeric, paprika and mustard seed powders each fluoresce quite well under UV light. However, these spices may change the flavor of produce if not sufficiently diluted. Similar effects can be achieved using synthetic dyes with and/or without color pigments. Those skilled in the art will appreciate that there are various other food safe fluorescing dyes suitable for use according to embodiments of the invention.
[0055] Referring now to Figure 5, illustrated is a schematic diagram of a first system 100 for defect identification according to an embodiment of the invention. The system 100 includes a controller 30 in managerial communication with a number of devices that wash, dry, organize, apply a food safe fluorescing dye and operate to identify pieces of produce that have unacceptable defects that are more easily detectable using UV light after the food safe fluorescing dye has been applied.
[0056] In some embodiments the controller 30 includes a computer readable program code means for identifying defects in produce. The computer readable program code means includes instructions for optically scanning produce under UV light; and, instructions for parsing scanned images of the produce under the UV light to determine whether or not the fluorescence emitted from the fluorescing dye is indicative of a defect or an acceptable natural feature of the scanned produce. Moreover, the further instructions for indicating that a piece of produce is defective when it has been determined that the fluorescence emitted from the fluorescing dye is indicative of a defect in the produce.
[0057] Specifically, the system 100 also includes a conveyor system having a first conveyor 51 , a second conveyor 53 and a third conveyor 55 arranged in series and operable to organize and move produce 41 through the system 100. The controller 30 preferably manages the speed of each of the conveyors 51 , 53 and 55. The system 100 also includes a water wash station 31 , an air dry station 33, a dye submersion tank 35, a water rinse station 37 and a set of UV lights and sensors 39. The air dry station 33 may be an arrangement of blowers and/or a distance over which the pieces of produce 41 dry in the open air en route to the dye submersion tank 35. The sensors (not specifically indicated) may include cameras and/or photosensors capable of detecting fluorescing dye illuminated by the UV lights. In this very specific example the water wash station 31, the air dry station 33, the dye submersion tank 35, the water rinse station 37 and the set of UV lights and sensors 39 are arranged in sequence along the series path of the conveyors 51 , 53 and 55. In particular, the water wash station 31 and the air dry station 33 are arranged one after the other over the first conveyor 51. The second conveyor 53 is positioned in the dye submersion tank 35 and angled upwards in the direction of travel to deliver produce 41 out of the dye submersion tank 35. The second conveyor 53 includes spacers 52 that separate pieces of produce 41 as they are drawn from the dye submersion tank 35 and delivered to the third conveyor 55. The water rinse station 37 and the UV lights and sensors 39 are positioned along the path of the third
conveyor 55. Similar to the second conveyor 53, the third conveyor also has spacers 52 that separate pieces of produce 41 from one another by a near- equal distance or multiple thereof.
[0058] In operation, pieces of produce 41 are delivered to the input end of the system 100, which is at the open end of conveyor 51. The pieces of produce 41 are washed and dried at the water wash station 31 and the air dry station 33, respectively. The pieces of produce 41 then fall into the dye submersion tank 53 where they are coated with the food safe fluorescing dye for a controllable duration. The average amount of time each piece of produce 41 spends submersed in the dye may be managed to ensure that the dye may sufficiently be absorbed so that the contrast between defects and acceptable natural features of the type of produce being inspected is suitably enhanced when examined under UV light. For example, when inspecting black plums, it is preferable to submerse the black plums in a solution of quinine sulphate for about 15 seconds for good results.
[0059] The spacers 52 on the second and third conveyors 53, 55 separate pieces of produce 41 from one another by a near-equal distance or multiple thereof, as the pieces of produce 41 are taken through water rinse station 37 and the UV lights and sensors station 39. Pieces of produce 41 that have defects that are identified by the UV lights and sensors station 39 can be removed and discarded.
[0060] Turning to Figure 6, illustrated is a schematic diagram of a second system 200 for defect identification according to another embodiment of the invention. The system 200 includes many of the same elements included in the system 100 illustrated in Figure 5, and, accordingly, common elements that are substantially the same in both systems 100, 200 have been designated using the same reference numbers.
[0061] Instead of a number of conveyors, the system 200 includes only one conveyor 61 that defines a path for the arrangement of other devices to be arranged along. Specifically, the system 200 includes the water wash station 31 , the air dry station 33, the water rinse station 37 and the set of UV
lights and sensors 39 arranged in sequence along the path defined by the conveyor 61. Moreover, in contrast to system 100, system 200 includes an overhead dye wash station 71 , instead of a dye submersion tank. The overhead dye wash station 71 is positioned between the air dry station 33 and the water rinse station 37, and is specifically placed a delay distance di in front of the water rinse station 37.
[0062] The operation of the system 200 is similar to the first system
100. The substantial difference lies in the application of the food-safe fluorescing dye, which, in the system 200 is applied by overhead sprayers and/or spillers to the pieces of produce 41. Moreover, while the pieces of produce travel from the overhead dye wash station 71 to the water rinse station 37, the delay distance d^ provides enough time for the food safe fluorescing dye to sufficiently absorb to ensure that the difference between any defects and the acceptable natural features of the produce is sufficiently enhanced. The time it takes for pieces of produce 41 to traverse the delay distance di is controllable by changing the speed of the conveyor 61 , which is preferably managed through the controller 30. In such embodiments, the controller 30 may include a computer program readable code means with instructions for controlling the speed of conveyor 61 (and/or various other conveyors included the system).
[0063] Referring to Figure 7, illustrated is a schematic diagram of a third system 300 for defect identification according to another embodiment of the invention. The third system 300 is the aggregation of a number of sorting lanes that are each controlled by a central controller 30. In fact, the system 300 may include the features and arrangements described above with respect to the first and/or second systems 100, 200 for each of three parallel sorting lanes A, B, C provided with reference numbers 81 , 83 and 85, respectively.
[0064] In operation, pieces of produce 41 arranged in lanes are delivered to the respective sorting lanes Lane A 81 , Lane B 83 and Lane C 85. In each sorting lane the pieces of produce 41 are dyed with the food safe fluorescing dye and, subsequently inspected for defects under UV light, as is
done in systems 100, 200. The pieces of produce 41 identified as having defects are rejected out of the sorting lanes, while acceptable produce 41 in each of the lanes is sent upstream for further processing.
[0065] Turning to Figure 8, illustrated is a schematic diagram of a fourth system 400 for defect identification according to another embodiment of the invention. The system 400 includes many of the same elements included in the system 100 illustrated in Figure 5, and, accordingly, common elements that are substantially the same in both systems 100, 400 have been designated using the same reference numbers. [0066] Specifically, the system 400 includes a conveyor system having the first conveyor 51 , the second conveyor 53 and the third conveyor 55 arranged in series and operable to organize and move produce 41 through the system 400. The controller 30 preferably manages the speed of each of the conveyors 51 , 53 and 55. The system 400 also has a dye submersion tank 35. However, instead of an automated unit containing UV lights and sensors, the third conveyor defines a manual sorting area 63 in which people inspect dye produce under UV light provided by UV lights 62 and 64.
[0067] While the above description provides example embodiments, it will be appreciated that the present invention is susceptible to modification and change without departing from the fair meaning and scope of the accompanying claims. Accordingly, what has been described is merely illustrative of the application of aspects of embodiments of the invention. Numerous modifications and variations of the present invention are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims, the invention may be practiced otherwise than as specifically described herein.
[0068] For example, existing produce sorting and defect identifying systems may be provided with a kit of parts suitable for upgrading a produce sorting system to identify produce defects using a combination of a food safe fluorescing dye and ultraviolet light in accordance with an embodiment of the invention. The kit of parts may include items such as an applicator for
applying a food safe fluorescing dye to pieces of produce; a UV light source for emitting UV light and thereby activating fluorescence from the food safe fluorescing dye; and a controller, connectable to receive an input from the UV light sensor to determine whether or not the fluorescence emitted from the fluorescing dye is indicative of a defect or an acceptable natural feature of the produce being inspected. Additionally and/or alternatively, the kit of parts may include a computer readable program code means for identifying defects in produce, the computer readable program code means having instructions for optically scanning produce under UV light; and, instructions for parsing scanned images of the produce under the UV light to determine whether or not the fluorescence emitted from the fluorescing dye is indicative of a defect or an acceptable natural feature of the produce being inspected.