EP4346374A1

EP4346374A1 - Method and device for automated inspection of plants and solid growth media

Info

Publication number: EP4346374A1
Application number: EP22727907.2A
Authority: EP
Inventors: Stephan VON RUNDSTEDT; Friederike Von Rundstedt
Original assignee: Robotec Ptc GmbH
Current assignee: Robotec Ptc GmbH
Priority date: 2021-05-26
Filing date: 2022-05-06
Publication date: 2024-04-10
Also published as: DE102021113510A1; WO2022248191A1

Abstract

The invention relates to a method and to a device by means of which the processing of plants can be made more efficient. The problem is solved by the plants (15) and in particular plant-based and artificial solid growth media or substrates being subjected to automated inspection. For this purpose, a sample and/or an image of at least one plant (15) or at least one solid growth medium or the substrate is automatically created by a sensor unit (20, 21). This sample and/or this image is then compared by means of an evaluation device with known samples and/or images of plants (15) and/or solid growth media/substrates which are contaminated.

Description

Method and device for the automated scoring of plants and culture media

description

The invention relates to a method for the automated scoring of plants and culture media according to claim 1. The invention also relates to a device for the automated scoring of plants and culture media according to claim 11.

Plant tissue culture has been used commercially for more than 40 years to vegetatively propagate high quality plant material. Large numbers of plants of consistent quality can be produced in sterile conditions on artificial culture media in culture vessels in short periods of time. In addition to mass propagation, plant tissue culture is also used for other purposes in plant breeding and, for example, for disease-free preservation and initial propagation of simpler plant species such as bedding and balcony plants.

Contaminations in plant tissue culture can come from two sources, namely microorganisms on the surface or in the tissue of the plant explants or from defective sterilization or processing procedures in the laboratory. Plant surfaces and tissues are habitats for microorganisms (mainly bacteria and fungi, but also mites, for example). During the course of plant growth and development, many surface and rhizosphere-dwelling microorganisms can opportunistically penetrate the plant tissues through natural openings, wounds, etc. and colonize them to varying degrees. In addition, facultative and obligate pathogens, whether vector-assisted or via other penetration mechanisms, can colonize and host plants in a similar manner. Plants can thus develop an endophytic "flora" (endogenous) of variable species composition, consisting of inter- and intracellular microorganisms including viruses, viroids, prokaryotes (bacterial and bacteria-like pathogens) and fungi. When establishing tissue cultures, depending on the explant used, surface and endophytic microorganisms can be introduced into the culture. Meristem culture eliminates most organisms, depending on meristem size, while leaf, petiole, or stem explants allow most, if not all, microorganisms to be transferred into the tissues.

For economic in vitro multiplication in a short time with high volumes, it is essential to keep the in vitro cultures in a balance of these microorganisms so that bacteria or fungi, for example, do not grow on the culture media. Each processing or propagation step is carried out under sterile conditions on the sterile workbench, since otherwise there is a risk of contaminating the in vitro cultures with exogenous microorganisms. After the subculture period of the plants to be propagated on the artificial culture medium under artificial conditions, the plants are processed again. Before this renewed processing, the plants or the culture vessels are subjected to a visual inspection (scoring) for contamination and quality, so that no contaminated and/or poor-quality plant cultures are propagated. For this rating, the plants are viewed individually by people. As a result, the process is not only very time-consuming but also expensive.

The invention is based on the object of creating a method and a device with which the processing of plants can be made more efficient.

A solution to this problem is described by the measures of claim 1. Accordingly, it is provided that the plants and in particular plant and artificial culture media or substrates are subjected to an automated assessment. For this purpose, a sample and/or an image of at least one plant or at least one nutrient medium or the substrate is automatically created from the plant or from the nutrient medium or the substrate by a sensor unit. This sample and/or this image is then compared by an evaluation device with known samples and/or photographs of plants and/or culture media/substrate that are contaminated. This automatic comparison of the plants or the culture media or the substrate with samples or recordings that show contamination or the like allows a visual inspection to be carried out quickly and reliably will. The time-consuming and costly deployment of people as well as incorrect visual measurements and counts can thus be avoided. The method described here is used in particular for processing useful and ornamental plants. However, it is equally conceivable that the claimed method can also be used for scoring other objects. The claimed plants can also be parts of plant organs or other plant material, tissue, cells or the like. Accordingly, the method described here should not be restricted to use on a plant, but rather also extend to the use of other organic forms. As an alternative to the culture medium claimed here, the method and the device can also be used for container contents or substrates.

In particular, the invention further provides that the at least one plant, in particular cells, tissue, plant organs, and/or the at least one culture medium or the substrate is supplied to the sensor unit in at least one container or on a tray prior to assessment, the Feeding is carried out manually by a person or automatically by a conveyor or a gripper arm, the container being fed by the gripper arm to the sensor unit in such a way that the sample and/or the image can be created in a particularly efficient manner. When feeding the container to the sensor unit manually, the person holds the container under the sensor unit for a sufficiently long period of time. The person can be shown when the sample or the image was created and how much of a sample or plant or the like is sorted out. It also documents which quality defects were detected. With automatic feeding, no further personnel are required. Here the feeding and removal of the containers is carried out and controlled by a control unit. It is also conceivable that the at least one plant, in particular cells, tissue, plant organs, and/or the at least one culture medium or the substrate is placed on a shelf or tray by the gripper arm before the assessment and is fed to the sensor unit or that the sensor unit is fed to the tray.

Preferably, it is further provided that an image recognition device from a container or from a plurality of containers, each containing a plant and/or a Have nutrient medium, takes a picture and based on this image, the gripper arm is automatically moved to a container to take the container and feed it to the sensor unit. The image recognition device thus recognizes individual containers from the large number of containers and can then pick them up one after the other in a particularly efficient manner. Alternatively, it is also conceivable that several containers are gripped at the same time and fed to the sensor unit, or that the sensor unit is fed to the containers and scans the individual containers.

Preferably, it is also conceivable that the individual containers are transported on a conveyor through a first lock into a room, in particular a sterile room, with the evaluation being carried out before entry into the first lock or only in the room. After processing in the room, the individual containers can then be transported out of the room again through a second lock. The lock can prevent any contamination or the like from being transported into the sterile interior. This supply and removal of the individual containers can be done both manually and automatically.

It is also conceivable that the individual containers are transported into the room in a closed state and are then opened manually or by a gripper arm and then the assessment is carried out in the room. For example, a robotic arm with a suction cup can be used to open the container, which lifts the lid of the container and places it back on the container after processing. During processing, the lid can be picked up in a multi-lid magazine, with each lid being reassigned to exactly the same container. Alternatively, it is also conceivable that new, in particular sterile, lids from a lid dispenser are used in order to close the container again.

Another particularly advantageous embodiment of the invention can provide for the sample and/or the recording to be created by at least one camera for visible, infrared and/or ultraviolet light, a pH value measuring device, an impedance spectroscope, a gas sensor or the like. A large number of measurements, in particular complementary measurements, can be carried out by these sensors in order to determine any contamination or infections of the plant or the culture medium or the substrate. Furthermore, the invention can provide that an evaluation of the created samples and/or the images is carried out by a neural network, with the neural network or the evaluation device being provided with a database with a large number of samples and/or images, which can be linked to contamination or infection of the plant or the medium/substrate. By using a neural network, the evaluation device can continuously learn and improve the detection of contamination. The database made available allows contamination or infections to be detected very quickly and reliably. This database can be expanded as required for each new type of plant or growing medium.

An additional exemplary embodiment of the invention can provide for several of the named sensor devices to be used simultaneously or one after the other in order to carry out the assessment, with the neural network determining in particular during the creation of the sample and/or the image whether a further sample and/or another picture is to be created. In particular in cases where the result is not clear or cannot be verified, it can be advantageous for a further measurement to be carried out in order to confirm the measurement result. This possibility of multiple use of the sensors reduces the rejection rate of unrecognized samples or images. Finally, it is also conceivable that in ambiguous or doubtful cases, individual plants, substrates or cups or tablets are additionally checked by a person in order to reintroduce them into the further processing chain or to discard them completely.

In addition, another aspect of the invention can be that based on the comparison of the created samples and / or images of the plant or the culture medium or a substrate with the stored on the database samples and / or images of the neural network or the artificial Intelligence determines whether the plant or the culture medium or the substrate is fed to further process steps, separated out or subjected to a special treatment. Due to this further aspect, the method described here consists not only in carrying out a visual inspection of the plant or the culture medium/substrate, but also in a further decision on how to proceed with the plant or the culture medium or the substrate after the inspection. This further process step allows the processing to be controlled by objective criteria. Whereas previously the person had to decide how to proceed with the plant or the culture medium or the substrate, this is now done uniformly on the basis of the neural network.

Finally, it is conceivable according to the invention for the evaluation device to generate a corresponding signal in the event of a detected contamination or infection of the plant or the culture medium/substrate, or for the container to be sorted out, preferably by a gripping arm. By selectively separating out the container, the contamination found is ejected from the process in a very effective manner, so that the remaining plants or growing media are protected accordingly. Early detection of contamination can protect a wide range of other plants and growing media, making the entire process more cost-effective.

Claim 11 describes a device for solving the above-mentioned problem. Accordingly, it is provided that the device for automated assessment of plants and in particular plant or artificial culture media has a sensor unit for automatically creating a sample and/or an image of at least one plant or at least one culture medium/substrate. In addition, the device has an evaluation device for the automated detection of contamination of the plant or the culture medium/substrate.

The sensor unit can be a camera that is sensitive to visible, infrared and/or ultraviolet light. While some infections or fungi can be seen directly, other contaminations can become visible in a different wavelength range. In particular, by irradiating the plant or the nutrient medium/substrate with light of a specific wavelength, spores or the like that would not be visible in the visible spectrum can be made visible by fluorescence. Hyperspectral cameras are particularly suitable for making such contaminations recognizable. To particular smallest To identify fungi, bacterial colonies, mites, etc., it is also conceivable to use a magnifying lens or a microscope with an imaging sensor unit. In addition, the sensor unit can be designed as a pH value measuring device. By changing the pH value, infections in the medium can be detected particularly early and precisely. For this purpose, a corresponding sample can be taken from the plant or the culture medium, which is immediately measured by the measuring device. In addition, it is conceivable that an impedance spectroscope can be used to determine other properties of the plant or the culture medium that indicate possible contamination or infection. A further embodiment can provide for the sensor unit to be in the form of a gas sensor. By determining the composition of the atmosphere and, if necessary, a change in composition, infections can be detected even if they are not yet visible. A particularly advantageous exemplary embodiment can also provide for the device to have several of the sensor units mentioned in order to carry out various measurements on the plant or the nutrient medium. The measurement result can be improved by a complementary measurement.

In addition, it is conceivable according to the invention that the evaluation device is based on artificial intelligence with a neural network, with the neural network having a database on which a large number of samples and/or images of plants and culture media with a contamination or an infection are stored . This artificial intelligence recognizes patterns and features of the individual images and samples and compares them with known data. The samples and images created in this way are used to enlarge the database and thus constantly improve the hit accuracy of the neural network.

Finally, it is provided that the device has at least one conveying device or one gripper arm, in particular a robot arm, through which a container in which the plant and/or the culture medium are placed can be fed to the sensor unit. This conveying device or the gripping arm is controlled by a control device which is connected to the neural network and is controlled on the basis of the sensor unit or image recognition device. This combination or integration of different components of the device makes it possible carry out prescribed procedures in a particularly time- and cost-efficient manner.

A possible embodiment of the invention is described in more detail below with reference to the single figure of the drawing.

1 shows a highly schematized device.

A possible exemplary embodiment of the device 10 according to the invention is shown in the figure. It is provided according to the invention that this device 10 can also be integrated into other devices not shown here and can represent part of a complex treatment process for plants.

In the exemplary embodiment illustrated in the figure, a plurality of containers 11 are transported on a conveyor device 12 in the conveying direction 13 . These containers 11 can be closed by a cover 14 . The containers 11 contain both a non-visible nutrient medium or a substrate and a plant 15 . The individual containers 11 can be placed on the conveying device 12 manually or automatically. The containers 11 can already have passed through a sluice (not shown) or are fed into a sluice in the further process in order to be further processed in a sterile room.

The device 10 also includes a control unit 16. In the exemplary embodiment shown here, this control unit 16 controls an image recognition device 17, a first gripper arm 18, a second gripper arm 19, a first sensor unit 20 and a second sensor unit 21. In addition, the conveyor device 12 can also be controlled via the control unit 16 . For the control of the aforementioned components, the control unit 16 has at least one processor and one evaluation device. In addition, the control unit 16 can also have a neural network that supports the method described here

First, the containers 11 transported on the conveyor device 12 are detected by the image recognition device 17 . The image recognition device 17 or the control unit 16 determines the type or size of the individual containers 11, the number of individual containers 11 and, if necessary, reads identification numbers or descriptions on the individual containers 11. The next process step is determined by the control unit 16 on the basis of this information. In the case of a container 11 closed by a lid 14 , this lid 14 is first lifted by the first gripping arm 18 . For the further process, this cover 14 can either remain on the gripping arm 18 or be supplied to a magazine (not shown). When the container 11 is subsequently transported further, the same cover 14 is preferably supplied to the same container 11 again. The gripping arm 18 has a corresponding gripping means 22 for handling the lid 14 . This gripping means 22 can be a suction cup, for example. This gripping means 22 is movably arranged on the robot arm-like gripping arm 18 and is also controlled via the control unit 16 . For this purpose, the gripping arm 18 can preferably be moved in three-dimensional space.

In a subsequent method step, the opened container 11 with the nutrient medium/substrate and/or the plant 15 is supplied to a sensor unit. In the exemplary embodiment shown here, the container 11 is supplied to two sensor units 20, 21. This feeding is carried out by the second gripping arm 19. This second gripping arm 19 is also designed like a robot arm and also has a gripping means 23. This gripping means 23 is designed in such a way that it can grip the container 11 and feed it precisely to a sensor unit. The container is then placed back on the conveyor 12 by the gripper arm 19 and closed with the lid 14 .

The sensor units 20, 21 shown in the figure are a camera that can take an image or a recording of the plant 15 or the culture medium. The second sensor unit 21 is a measuring device for determining the pH value of the culture medium. Equally, however, it is also conceivable for the device 10 to have only one sensor unit or further sensor units. These complementary measurements allow both the plant 15 and the culture medium to be assessed.

The information recorded by the sensor unit 20, 21 is evaluated by the control unit 16. The neural network can be used here, with the neural network using the recorded information or the Recordings and the samples are compared with stored samples. The stored patterns can represent, for example, recordings or samples of corresponding plants or culture media that show contamination or an infection. By comparing the known data with the data created by the sensor units 20, 21, the control unit 16 can carry out the rating. The further treatment of the plant 15 or the culture medium takes place depending on the result of the assessment.

Due to the constant pattern recognition by the neural network, the network is permanently trained so that the probability of hits is improved with each comparison carried out. By using the neural network, different plants or different culture media can also be flexibly assessed, since the neural network recognizes the type of plant or culture media involved. Depending on the recognized plant or the culture medium, the neural network can access different data sets.

As an alternative to the exemplary embodiment shown in the figure, it is also conceivable for a person to open the container 11, feed it to the sensor unit 20, 21 and then set it down, for example, on a conveyor (not shown). It is also conceivable for a person to support the method presented here by confirming the contamination recognized by the neural network or the artificial intelligence or evaluating it as incorrect in order to train the neural network even further.

A further aspect of the invention could consist in the control unit 16 controlling further devices which immediately treat the plant 15 or the culture medium, the substrate, the container and/or the tray after an infection has been detected. For example, the opened container 11 could be subjected to a short UV, UVC or X-ray flash, chemical spraying or gassing with H2O2, chlorine dioxide, NaOCl, a surface disinfectant or the like in order to kill any germs or the like that were found. It is also conceivable to re-sterilize plants, substrates or cups in the event of animal contamination by means of CO ₂ pressure. If required, the container 11 could also are immediately fed to a conveyor, not shown, for separating out the plant 15 or the culture medium.

May 5, 2022

Reference List

10 device

11 containers

12 conveyor

13 conveying direction

14 lids

15 plant

16 control unit

17 image recognition device

18 first gripper arm

19 second gripper arm

20 first sensor unit

21 second sensor unit

22 gripping means

23 gripping means

Claims

patent claims

1. A method for the automated assessment of plants (15) and in particular plant or artificial culture media or a substrate, with a sensor unit (20, 21) automatically taking an image and/or a sample of at least one plant (15) or at least one culture medium or one Substrate is created and this sample and / or this image are compared by an evaluation device with known samples and / or recordings of plants (15) and / or culture media or substrates that are contaminated.

2. The method according to claim 1, characterized in that the at least one plant (15) and/or the at least one culture medium or the substrate is supplied to the sensor unit (20, 21) in at least one container (11) or on a tray, wherein the feeding is carried out manually by a person or automatically by a conveyor or a gripping arm (19), the container (11) or the tray being fed by the gripping arm (19) to the sensor unit (20, 21) in such a way that the sample and /or the image can be created in a particularly efficient way.

3. The method according to claim 1 or 2, characterized in that an image recognition device (17) from a container (11) or from a plurality of containers (11), each having a plant (15) and / or a nutrient medium or a substrate , takes an image and based on this image the gripper arm (19) is automatically moved up to a container (11) in order to grip the container (11) and feed it to the sensor unit (20, 21).

4. The method according to any one of the preceding claims, characterized in that the individual containers (11) are transported on a conveying device (12) through a first lock into an, in particular sterile, room, with the scoring before entering the first lock or in is carried out in the room.

5. The method according to claim 4, characterized in that the individual containers (11) are transported in a closed state into the room and then be opened manually or by a gripper arm (18) and the rating is then carried out in the room.

6. The method according to any one of the preceding claims, characterized in that the sample and/or the recording of at least one camera (20) for visible, infrared and/or ultraviolet light, in particular using a microscope, a pH value measuring device ( 21), an impedance spectroscope, a gas sensor or the like.

7. The method as claimed in one of the preceding claims, characterized in that the samples and/or images produced are evaluated by an artificial intelligence (KI), in particular by a neural network, with the AI or the evaluation device being provided with a database with a variety of samples and/or images that can be associated with contamination or infection of the plant (15) or the culture medium or the substrate.

8. The method according to any one of the preceding claims, characterized in that several of said sensor devices (20, 21) are used simultaneously or one after the other to carry out the evaluation, in particular during the creation of the sample and/or the image by the AI or the neuronal network it is determined whether another sample and/or another image is to be created.

9. The method according to any one of the preceding claims, characterized in that on the basis of the comparison of the created samples and / or images of the plant (15) or the nutrient medium or the substrate with the stored on the database samples and / or images of the neuronal Network is determined whether the plant (15) or the culture medium supplied to further process steps, separated or subjected to a separate treatment.

10. The method according to any one of the preceding claims, characterized in that by the evaluation device in the event of a detected contamination or infection of the plant (15) or the culture medium or the Substrate a corresponding signal is generated or the container (11), preferably by a gripper arm (19), is separated.

11. Device (10) for the automated assessment of plants (15) and in particular plant or artificial culture media or substrates with a sensor unit (20, 21) for automatically creating an image and/or a sample of at least one plant (15) or at least one culture medium or a substrate and an evaluation device for the automated detection of contamination on the plant (15) or the culture medium or the substrate.

12. Device (10) according to claim 11, characterized in that the sensor unit is at least one camera (20), in particular in connection with a microscope, for visible, infrared and/or ultraviolet light, a pH measuring device (21), an impedance spectroscope, a gas sensor or the like.

13. Device (10) according to Claim 11 or 12, characterized in that the evaluation device is based on artificial intelligence with a neural network, the neural network having a database on which a large number of samples and/or images of plants (15 ) or culture media or substrate with contamination or infection.

14. Device (10) according to any one of claims 11 to 13, characterized in that the device (10) has at least one conveyor device (12) or a gripper arm (19), in particular a robot arm, through which a container (11) or a Tray in which or on which the plant (15) and/or the culture medium are placed, can be fed to the sensor unit (20, 21).