WO2021161341A1

WO2021161341A1 - System for modular multi-grade sorting and a method thereof

Info

Publication number: WO2021161341A1
Application number: PCT/IN2021/050135
Authority: WO
Inventors: Gaurav PARDESHI; Kshitij THAKUR; Nikhil PANDEY; Rakesh BARAI
Original assignee: Occipital Technologies Private Limited
Priority date: 2020-02-14
Filing date: 2021-02-12
Publication date: 2021-08-19

Abstract

The present invention discloses a system and method for modular, multi-grade sorting of objects (108) including but not limited to, fruits, vegetables, seafood, and meats. The objects (108) are fed from an inclined feeder unit (102) through a segregator roller (300) onto a conveyor unit (110), which controls the flow of the objects (108). The objects pass through an inspection unit (112) comprises a processing unit (202), sensors and vision cameras to collect images and object information, in order to sort each object by grades and detect defects. The actuators (116) and sorting unit (114) are in sync with the inspection unit (112) so that the sorted objects (108) can be sorted automatically irrespective of the number of feeder units (102) or output bins used in the sorting system.

Description

System for modular multi-grade sorting and a method thereof

The field of the invention generally relates to a field of produce sorting machines. More specifically, it relates to a system and method for a modular, multi-grade produce sorting system.

Inspection and grading of food produce are a major issue for large-scale produce packaging industries. Currently, existing systems use cameras to capture images and analyze them for multiple parameters to classify those objects. However, the disadvantage of these systems is that they are able to perform only binary sorting. Hence, objects can only be segregated into two bins, namely, accepted or rejected bins. Additionally, such systems have complex conveying mechanisms.

Other existing systems have tried to address this problem. However, their scope was limited to non-modular designs which could only sort a single type of product or a limited range of similar products. Additionally, none of the current systems can accurately detect defects in the objects.

Thus, in light of the above discussion, it is implied that there is a need for a system which is reliable and does not suffer from the problems discussed above.

The present invention discloses a system for modular multi-grade sorting and a method thereof. The multi-grade sorting system grades and sorts objects that pass through the sorting system. The sorting system comprises a feeder unit into which the objects are placed. The feeder unit feeds the objects to a cleated conveyor belt, which comprises an inspection unit comprising various sensors. The objects pass into the inspection unit where the sensors measure and collect various information related to each object. The inspection unit comprises a processing unit which analyses the collected object information to determine an appropriate output bin into which the object needs to be sorted. Subsequently, the processing unit generates a unique signal based on the determined bin, for each object. Thereafter, the objects are passed on to a sorting unit, which uses the signals generated by the processing unit to sort each object into their appropriate bin.

The inspection unit further comprises a vision camera that is capable of detecting external and internal qualities of each object, as well as multiple sensors capable of detecting one or more parameters including but not limited to moisture, chemical composition, etc., related to the objects.

Further, the multi-grade sorting system comprises an actuator that is capable of actuating various sorting arms to move each object into corresponding bins.

Object of Invention

The principal object of this invention is to provide a system and method for multigrade sorting of objects.

Another object of the invention is to provide a system and method for multi-grade optical sorting which can sort objects into desired number of grades with easy modifications.

Yet another object of the invention is to provide a system which allows for instant changes to grade specifications by using a computer software.

Another object of the invention is to provide a minimal-friction isolation mechanism for sorting products with sensitive or delicate surfaces.

Another object of the invention is to provide a sorting system with a modular design which can be easily customized to vary the feeding or sorting section.

Another object of the invention is to provide a sorting system with automatic synchronization of modular components.

Another object of the invention is to provide a sorting system with a modular design which can easily integrate multiple sensors.

Another object of the invention is to provide a sorting system with image processing and machine learning models for weight estimation.

Another object of the invention is to provide an optical grading and sorting system to detect and segregate products affected by diseases and

This invention is illustrated in the accompanying drawings, throughout which, like reference letters indicate corresponding parts in the various figures.

The embodiments herein will be better understood from the following description with reference to the drawings, in which:

Fig. 1A

depicts/illustrates a block diagram of a modular multi-grade sorting system, in accordance with an embodiment;

Fig. 1B

illustrates the multi-grade optical sorting system, in accordance with an embodiment;

Fig. 2

depicts/illustrates an inspection unit which comprises a processing unit and vision camera, in accordance with an embodiment;

Fig. 3

depicts/illustrates a segregator roller used to simultaneously pick several objects of the same size range. in accordance with an embodiment;

Fig. 4

depicts/illustrates a stabilizer roller used to control flow of objects through the sorting system, in accordance with an embodiment;

Fig. 5

shows depicts/illustrates a segregator control wheel for allowing individual objects of a particular size at a time, in accordance with an embodiment;

Fig. 6

shows depicts/illustrates a push arm for pushing objects into a specific bin, in accordance with an embodiment;

Fig. 7

shows depicts/illustrates the modularity of a multi-grade sorting system for a large throughput requirement, in accordance with an embodiment.

The embodiments herein and the various features and advantageous details thereof are explained more fully with reference to the non-limiting embodiments that are illustrated in the accompanying drawings and/or detailed in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein. The examples used herein are intended merely to facilitate an understanding of ways in which the embodiments herein may be practiced and to further enable those of skill in the art to practice the embodiments herein. Accordingly, the examples should not be construed as limiting the scope of the embodiments herein.

The present invention discloses a modular multi-grade sorting system for sorting objects including, but not limited to, fruits, vegetables, seafood/meats and other food products. The objects are fed into the feeder unit comprising an inclined feeder which vibrates to aid the flow of objects onto the conveyor unit. The inclination of the feeder is designed such that the objects 108 can easily be transported onto the conveyor unit. The feeder unit also comprises a dampener and a segregator roller which allow the objects to pass onto the conveyor belt, one at a time. The feeder unit further comprises a segregator roller which decreases the load on the control wheel and allows only certain quantities of the objects of same size at the same time to move towards the control wheel.

The objects are transported on a cleated conveyor belt and passed through an inspection unit comprising a processing unit, multiple sensors and vision cameras. The inspection unit measures and collects various images and data related to each object on the cleated conveyor belt. The object data is analyzed in the processing unit by using a combination of image processing algorithms and machine learning models. Thereafter, the processing unit segments each object from an image background of the object and uses the machine learning models for grade or pattern detection and classification, in order to segregate the objects as per the predetermined bins.

The analyzed results of the inspection unit are in the form of voltage signals, which are fed to a voltage level converter or level shifter to shift the voltage to an input terminal of PLC for processing as per the algorithm. The output signal is then fed to a relay board to actuate the actuators as per the analyzed signals in a synchronized manner.

The appropriate signals are fed to the actuator which actuates the sorting arm to push the object into a desired passageway.

depicts/illustrates a block diagram of a modular multi-grade sorting system 100, in accordance with an embodiment. The sorting system 100 comprises a feeder unit 102, a conveyor unit 110, an inspection unit 112, and a sorting unit 114.

In an embodiment, objects 108 are fed into the feeder unit 102. The feeder unit ensures that the objects 108 are transferred to the conveyor unit 110 one at a time. Thereafter, the conveyor unit 110 transports the objects 108 to the inspection unit for sensing and collecting one or more object information from each object 108. Based on the information collected and analyzed by the inspection unit 112, the sorting unit 114 sorts each object such that it is pushed into an appropriate output bin.

In an embodiment, the feeder unit 102 comprises one or more flow control components such as a dampener 104, stabilizer, a control wheel 106, and a segregator roller (depicted in ). The mentioned components are used to control the flow of the objects 108 such that they are transferred onto the conveyor unit 110 one after another, such that the inspection unit can accurately collect and analyze each object 108.

In an embodiment, the inspection unit 112 comprises vision cameras, sensors and a processing unit. The vision cameras comprise optical sensors used to capture specific images of each object 108. The sensors are used to collect one or more object information comprising one or more of: NIR sensors, hyperspectral cameras for internal quality detection, spectrometers for detecting moisture and chemical composition, among others. The processing unit is used to analyze the collected images and object information by using image processing algorithms and machine learning models. After analyzing the images and object information, the processing unit grades each object on the basis of user defined criteria of physical parameters such as but not limited to, size, color, defect and determines which output bin each object needs to be sorted into. Based on the grade, the processing unit sends an appropriate output signal to the sorting unit 114. Thereafter, the sorting unit 114 performs the sorting action by using the output signal to trigger the actuators. The actuators push each object into its determined output bin.

In an embodiment, the sorting unit 114 can sort the objects 108 into a predetermined number of grades in a single pass, i.e., multiple grades of objects 108 can be sorted in only one pass of the objects 108 through the sorting system 100. A certain number of these sorted objects 108 can be stored in a storage area present at each output bin.

In an embodiment, the conveyor unit 110 comprises a cleated conveyor belt which enables the individual flow of products. The cleated design of the cleated conveyor belt 110 enables even distribution of objects along the conveyor belt 110, such that at any given time there is only one product between two consecutive cleats. This allows scanning of individual objects as the cleated conveyor belt restricts the motion of the objects 108 due to a low moment of inertia and acceleration of the object. The pitch of the cleated conveyor belt is designed as per the object’s dimensions and shape. The cleats perform the function of isolating each product and triggering one or more sensors placed along the cleated conveyor belt 110. The cleated conveyor belt isolates the objects 108 without rubbing against the entire surface of the object. Thus, the objects 108 can be sorted without any damage. Further, the triggers generated by sensors along the conveyor belt 110 can help in measuring and monitoring the speed of the cleated conveyor belt. Further, making changes in the speed of the cleated conveyor belt 110 can help in the synchronization of the sorting system 100.

illustrates the modular multi-grade sorting system, in accordance with an embodiment.

In an embodiment, the automated feeder unit 102 comprises an inclined feeder with inclined walls on either side, a dampener 104 and a control wheel 106. In an embodiment, the feeder is tilted at a particular inclination to allow the objects 108 to be easily fed into the feeder unit 102 through the control wheel 106.

In an embodiment, the dampener 104 works as a flow control component which reduces the load on the control wheel 106. In another embodiment, the control wheel 106 is designed specifically to let individual objects 108 pass at a time from the feeder unit 102 onto the conveyor unit 110.

depicts/illustrates an inspection unit 112 which comprises a processing unit 202 and vision camera 204.

In an embodiment, the vision camera 204 captures a specific number of images of the objects 108 under controlled illumination. The illumination is controlled by a processing unit which provides any required wavelengths or spectrum set by the user. The illumination range includes but is not restricted to ultraviolet, visible and infrared spectrum.

In an embodiment, the processing unit 202 uses a combination of image processing algorithms to pre-process the images, after which computer vision algorithms segment the object 108 from the image background and run machine learning models for grade/pattern detection and classification. The algorithms provide quality parameters such as but not limited to, dimensions, percentage of surface area covered by a particular defect, shape and colour at the output. The object 108 is classified into grades as per the user settings and output voltage signals are generated according to each object’s determined grade.

In an embodiment, the generated voltage signal of the inspection unit 112 is fed to a voltage level converter or level shifter to shift the voltage to a desired level for PLC for further processing. The output signal is provided to a relay board to enable actuators 116 to function as per the output signal, in a synchronized manner. The synchronized signal pulse is processed by a control panel and shared with the actuator 116. When the specific object 108 comes near the specific actuator 116, the actuator instructs the appropriate sorting arm 118 to move the object 118 into a desired output bin or conveyor. The sorting arm 118 comprises mechanisms such as but not limited to, linear push arm, overhead swivel arm, one side swivel arm, cart wheel type swivel arm.

In an embodiment, the sorting system 100 comprises machine learning models for weight estimation. The data analyzed by the image processing algorithms is used to estimate the weight of a particular product without using any additional sensors.

In an embodiment, the sorting system 100 comprises an easy to use user interface which allows instant changeable grade specifications. The software in the processing unit allows for instant changes to grade specifications allotted to each sorting line. The software can also aid in partial consideration of defects as per user requirements. In an embodiment, the mentioned user interface can be present in one or more of a server, a machine, a user’s device, or a smartphone app which can communicate with the server or the machine.

depicts/illustrates a segregator roller 300 with specially designed teeth used to pick several objects 108 in the same size range at a time. The objects are picked out such that multiple similar sized objects 108 can be selected for transporting to conveyor unit 110.

In an embodiment, each row of teeth on the segregator roller 300 can be of varying lengths. Additionally, inter-teeth distance can also vary for each row of teeth.

depicts/illustrates a stabilizer roller 400 used to control flow of objects 108 through the sorting system 100, in accordance with an embodiment.

The stabilizer roller 400 is used to control the flow of the objects 108, so that the control wheel 106 is able to allow the objects 108 to pass individually on to the conveyor unit 110.

depicts/illustrates a segregator control wheel 500 for allowing individual objects 108 of a particular size to pass onto the conveyor unit 110 at a time, in accordance with an embodiment.

depicts/illustrates a push arm 600 for pushing specific objects 108 into a specific bin, in accordance with an embodiment. The push arm 600 is placed at various points along the conveyor belt 110 and is used to push specific objects 108, by changing their trajectory. The objects 108 may drop into a specific bin or conveyor belt, in order to be sorted out from the rest of the objects 108.

depicts/illustrates the modularity of the multi-grade sorting system 100 for a large throughput requirement, in accordance with an embodiment. In an embodiment, multiple lanes for input and output can be added to the sorting system 100, in order to increase the number of grades of objects 108 sorted by the sorting system 100.

The present system discloses a customizable, modular design which allows the user to add multiple bulk conveyors 702/1, 702/2, and 702/3 and multiple cleated conveyor belts 110/1, 110/2, and 110/3 to the sorting system 100. In an embodiment, the segregator control wheels 500/1, 500/2, and 500/3 allow the objects 108 to pass one after another onto the conveyor units 110/1, 110/2, and 110/3.

In an embodiment, the objects 108 are fed in bulk into the bulk feeders 702/1, 702/2, and 702/3, where each bulk feeder 702 comprises a segregator roller 300.

In an embodiment, the segregator rollers 300 on the bulk feeders 702/1, 702/2, and 702/3 comprise specially designed unique teeth which pick several objects 108 in the same size range to pass them onto the feeders 102/1, 102/2, and 102/3 respectively.

In an embodiment, the feeder unit 102 can be constructed as depicted, with inclination that aids the movement of the objects 108 towards the segregator control wheels 500/1, 500/2, and 500/3. This flow is controlled by the respective stabilizer rollers 400.

In an embodiment, the feeder unit 102 can comprise two or more conveyors assembled in a “V” shape, where the conveyors are run at different speeds to help in aiding the objects to flow in a single line.

In an embodiment, the actuator, conveyor unit and inspection unit are in complete synchronization with the control panel. The synchronization ensures that the speed of actuation is calculated automatically within the control panel, depending upon the one or more of speed of the cleated conveyor belt, the dimensions of cleats on this cleated conveyor belt and the type of objects being conveyed.

In an embodiment, the sorting system 100 can grade and sort objects into multiple grades as per one or more of shape, size, surface quality, and color. Therefore, the sorting system 100 can be used for multiple types of commodities/objects having similar physical properties. Based on the number of types/grades of commodities, parameters of the system such as the speed of conveyor system, motion of sorting arm and the cleated conveyor belt can be determined and automatically synchronized by the sorting system 100.

The present system provides various advantages as follows: Modular design: the modular design allows easy upgrades to the number of grades which can be increased or decreased by adding or removing actuators from the sorting system. Additionally, as the actuator, conveyor system and inspection unit are in complete synchronization with the control panel, the system provides an easy, hassle-free solution for achieving modularity.

Another advantage of the multi-grade optical sorting system 100 is that the design of the system makes it possible to add multiple cameras and sensors to collect additional data of the objects 108. Additionally, the system can be used for detecting defects in objects 108.

A further advantage of the present invention is that the cleated conveyor belt system restricts the flow of the objects 108 to one at a time which in turn makes it possible to scan individual objects 108 by restricting their motion. Another advantage is that such a system requires low energy for isolating each and every object compared to rolling systems.

The foregoing description of the specific embodiments will so fully reveal the general nature of the embodiments herein that others can, by applying current knowledge, readily modify and/or adapt for various applications such specific embodiments without departing from the generic concept, and, therefore, such adaptations and modifications should and are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. Therefore, while the embodiments herein have been described in terms of preferred embodiments, those skilled in the art will recognize that the embodiments herein can be practiced with modification within the scope of the embodiments as described here.

Claims

A modular multi-grade sorting system (100) comprising:
at least one feeder unit (102) to individually feed objects (108) to a conveyor unit (110);
the conveyor unit (110) for transporting the objects (108) through the sorting system (100);
an inspection unit (112) comprising a processing unit (202) and one or more vision cameras (204) for collecting one or more object information from each of the objects (108) to generate output signals for each of the objects (108); and
a sorting unit (114) configured to sort each of the objects (108) into an appropriate output bin based on the output signals generated by the inspection unit (112).
The modular multi-grade sorting system (100) as claimed in claim 1, wherein said conveyor unit (110) comprises a cleated conveyor belt to allow individual flow of the objects (108) through the sorting system (100), wherein pitch of said cleated conveyor belt is varied as per dimensions and shape of the objects (108).
The modular multi-grade sorting system (100) as claimed in claim 2, wherein one or more cleats on the cleated conveyor belt trigger sensors positioned along the conveyor unit (110) which measure speed of said conveyor unit (110).
The modular multi-grade sorting system (100) as claimed in claim 1, wherein the sorting system (100) is modular, and wherein the sorting system (100) comprises multiple bulk conveyors (702) and multiple conveyor belts (110) that are configured to be synchronized for smooth flow of objects through the sorting system (100).
The modular multi-grade sorting system (100) as claimed in claim 1, wherein said one or more flow control components of the at least one feeder unit (102) comprise:
a dampener (104);
a segregator roller (300) comprising multiple rows of teeth to simultaneously pick same-sized objects (108), wherein each of the multiple rows of teeth is of varying length, and wherein inter-teeth distance varies for each of the multiple rows of teeth;
a control wheel (106) designed to individually transport each of the objects (108) from said feeder unit (102) onto said conveyor unit (110); and
a stabilizer roller (400) designed to control flow of the objects (108) towards said conveyor unit (110).
The modular multi-grade sorting system (100) as claimed in claim 1, wherein said inspection unit (112) comprises:
one or more vision cameras (204) for capturing images of each of the objects (108);
multiple sensors for detecting one or more parameters comprising moisture, chemical composition, shape, colour and defects of the objects (108); and
a processing unit (202) configured to:
analyze the collected object information;
grade each of the objects (108);
determine an appropriate output bin for each of the objects (108) based on the grade; and
generate an output voltage signal based on the determined output bin for each of the objects (108).
The modular multi-grade sorting system (100) as claimed in claim 6, wherein said inspection unit (112) comprises one or more of cameras, NIR sensors, hyperspectral cameras, and spectrometers.
The modular multi-grade sorting system (100) as claimed in claim 1, wherein the processing unit (202) controls illumination of the objects (108) during the collecting of the object information, wherein the processing unit (202) varies the illumination comprising a range of one or more of ultraviolet, visible and infrared spectrum, based on user input.
The modular multi-grade sorting system (100) as claimed in claim 1, wherein said processing unit (202) utilizes a combination of image processing algorithms and machine learning models.
The modular multi-grade sorting system (100) as claimed in claim 1, wherein said processing unit (202) grades each of the objects (108) based on one or more user-defined criteria of physical parameters comprising size, color and defect.
The modular multi-grade sorting system (100) as claimed in claim 1, wherein said sorting unit (114) comprises:
at least one actuator (116) actuated by a control panel based on the output signals from said inspection unit (112), and
at least one sorting arm (118) triggered by said at least one actuator (116) to push each of the objects (108) into the appropriate output bin.
A method for modular multi-grade sorting of objects (108), the method comprising:
feeding the objects (108) individually from at least one feeder unit (102) to a conveyor unit (110);
transporting the objects (108) through a sorting system (100) by using the conveyor unit (110);
collecting one or more object information from each of the objects (108);
generating output signals for each of the objects (108) based on the collected object information by using an inspection unit (112); and
sorting each of the objects (108) into an appropriate output bin based on the generated output signals, by using a sorting unit (114).
The method for modular multi-grade sorting as claimed in claim 12, wherein the feeding of the objects (108) to the conveyor unit (110) comprises:
feeding into inclined feeder units (102);
simultaneously picking same-sized objects (108) by using a segregator roller (300);
individually transporting each of the objects (108) by using a control wheel (106); and
controlling flow of the objects (108) by using a stabilizer roller (400).
The method for modular multi-grade sorting as claimed in claim 12, wherein transporting the objects (108) comprises:
comprising transporting said objects (108) from said feeder units (102) onto conveyor units (110) comprising cleated conveyor belts, wherein pitch of said cleated conveyor belt is varied as per dimensions and shape of the objects (108).
The method for modular multi-grade sorting as claimed in claim 12, wherein the generating of the output signal comprises:
varying illumination of the objects (108) during the collecting of the object information based on user input;
capturing specific images of each of the objects (108) to detect external and internal qualities, by using multiple vision cameras (204);
detecting one or more parameters comprising moisture, chemical composition, shape, colour and defects of the objects (108) by using multiple sensors;
analyzing the images and data of each of the objects (108) with image processing algorithms and machine learning models by using a processing unit (202);
grading each of the objects (108) based on one or more user-defined criteria of physical parameters comprising size, color and defect;
determining an appropriate output bin for each of the objects (108); and
generating an output voltage signal based on the determined output bin for each of the objects (108) by using the processing unit (202);
The method for modular multi-grade sorting as claimed in claim 12, wherein the sorting comprises:
actuating at least one actuator (116) based on received output signals from said inspection unit (112); and
triggering at least one sorting arm (118) to push each of the objects (108) into the appropriate output bin, by using the at least one actuator (116).