Method for operating a container inspection system and control device for a container inspection device

Abstract

The invention relates to a method for operating a container treatment system for treating container parts for containers, plastic containers and / or bottles, in which a transport device transports the container parts as a container part stream from one treatment device of the container treatment system to another treatment device of the container treatment system along a transport path, wherein a sensor device detects, in particular optically, sensor data relating to the container parts, in particular spatially resolved sensor data and camera images, in order to carry out a container inspection task, and a real-time evaluation device evaluates the spatially resolved sensor data in real time by means of a machine learning container inspection model, the model comprises a set of parameters set as values learned as a result of a machine learning method, characterized in that a set of container part features based on the machine learning method is predetermined, and on the basis of the predetermined set of container part features, a system inspection task different from the container inspection task is determined. The spatially resolved sensor data acquired are evaluated, at least one system inspection variable being determined as a function of an inspection result of the system inspection task performed, the at least one system inspection variable being provided for controlling and / or regulating the container treatment system.

Description

[0001] This invention relates to a method for operating a container handling system for processing multiple container components for containers, preferably plastic containers and / or bottles; a control device for the container handling system; and a container handling system. A transport device is provided therein, which transports multiple container components as a flow of container components along a predetermined transport path.

[0003] Preferably, the container is a plastic container (especially a PET container), a container whose main component is pulp, and / or a glass container and / or a metal can. The container can be used in the beverage and / or food and / or cosmetics and / or pharmaceutical industries.

[0004] A method for performing a setup operation of a container inspection apparatus is known from DE 10 2021 133 164 B3. In this method, a sensor device acquires spatially resolved sensor data related to a container to be inspected, and a real-time evaluation device evaluates the spatially resolved sensor data of a single container in real time using a settable real-time container inspection model. Furthermore, multiple spatially resolved sensor data sets are provided on a non-volatile storage device. During the setup operation, the setup device retrieves the stored multiple spatially resolved sensor data sets and evaluates a test container inspection model based on the retrieved multiple spatially resolved sensor data sets.

[0005] Here, the test container inspection model is evaluated based on the rejection rate according to the retrieved multiple spatially resolved sensor data.

[0006] The evaluation of whether to remove containers associated with the collected sensor data from the container component flow depends primarily on a predetermined threshold parameter that indicates when to remove containers exhibiting defective behavior. In this type of evaluation, containers that are not entirely without defects, but whose defective behavior is insufficient to warrant removal, are classified as "defect-free".

[0007] Here, as the applicant is aware of within the framework of this invention, the valuable information that can be provided by those containers that were not eliminated and were not entirely without defects is ignored.

[0008] The purpose of this invention is to overcome the known disadvantages in the prior art and provide a method for operating a container handling system for processing multiple container components for containers; a container handling system; and a control device for the container handling system, which provide an improved, finely configurable operation for high-quality processing of the container components.

[0009] According to the invention, this objective is achieved by the subject matter of the independent claims. Advantageous embodiments and further improvements of the invention are the subject matter of the dependent claims.

[0010] In a method according to the invention for operating a container processing system to process multiple container components for containers, preferably for plastic containers and / or bottles, a transport device transports multiple container components as a flow of container components along a predetermined transport path, preferably from and / or to at least one processing device of the container processing system, and particularly preferably from at least one processing device of the container processing system to at least another processing device of the container processing system (for performing at least one processing step on the multiple container components).

[0011] Preferably, the container is a plastic container (especially a PET container), a container whose main component is pulp, and / or a glass container and / or a metal can. The container can be used in the beverage and / or food and / or cosmetics and / or pharmaceutical industries. Examples include metal cans or bottles, such as glass bottles, pulp bottles, and plastic bottles.

[0012] "Container components for containers" can also be understood as the container itself. Therefore, a container component may be, for example, a preform that is produced into a fully formed container through a molding process, or it may be a fully formed container.

[0013] Container components for containers can also be container fittings, such as (preferably resealable) container closures (e.g., screw caps or bottle caps), (PET and / or plastic) caps, labels, (laser or direct printing) markings, fillers and / or (finished) container packaging, container assemblies or the like (and combinations thereof).

[0014] Therefore, the transport equipment can be, for example, a feed section (e.g., a feed rail) for the container sealing section, which transports the container sealing section from the collection equipment to the sealing equipment for sealing the container.

[0015] Preferably, the container component is an object that can be transported as a single unit by the transport equipment and transported independently of other transported container components.

[0016] Preferably, at least one (first) processing device and / or at least one other processing device (particularly in each case) performs at least one processing step on the container component (and preferably on the container) and / or on multiple container components.

[0017] Specifically, at least one processing step is performed on each of the multiple container components.

[0018] The processing steps of at least one (first) processing device and / or (at least one) other processing device may be selected from a set of processing steps, which include: an injection molding process for producing injection molded parts (e.g., plastic preforms); a cleaning process; a crushing process and / or a slitting process (e.g., within the framework of a recycling process); a molding process (particularly (stretch) blow molding); a (laser) marking process; a personalization process (e.g., affixing QR codes); a rejection process (in which container parts are rejected from the container part stream of the container handling system); and a sorting process (in which container parts are sorted according to type). The process includes: sorting container components; filling; sealing; (especially) direct printing; labeling; laser decoration; bundling (for container components, preferably containers); packaging (especially applying primary and / or secondary packaging, such as heat-sealing several containers into packages and / or bundling units), especially heat shrink packaging; determining the composition of a substance or mixture of substances, for example, by means of mass spectrometry and / or odor sensor equipment, to determine, for example, the composition of the gaseous environment and / or air within or at a container component, preferably within or at a container; and similar processes and combinations thereof.

[0019] To perform at least one container inspection task, at least one sensor device acquires sensor data, particularly spatially resolved sensor data, and preferably acquires camera images, preferably optical images, related to the container components (to be inspected and / or transported), (preferably during the operation of the container handling system). The sensor device is preferably a sensor device of the container handling system and / or a container inspection apparatus, which will be described in more detail below.

[0020] Preferably, the transport equipment transports multiple (to be processed and / or processed) container components to a sensor device (for inspecting the container components to perform a container inspection task), which in each case (particularly individually) acquires sensor data, particularly spatially resolved sensor data, related to the transported container component. The sensor device and / or a container inspection apparatus including at least one of the sensor devices (described in more detail below) may be arranged between the processing equipment and the at least one other processing equipment.

[0021] Preferably, at least certain areas of the container component (to be inspected), preferably areas of the container component that can be observed or seen from at least one viewing direction, are mapped into the sensor data.

[0022] It is conceivable that, for each container component to be inspected and / or transported to the sensor device, sensor data is acquired or collected individually or on a per-container-component basis (in a separate acquisition step for each of the sensor devices).

[0023] A single sensor device can be set up to collect the necessary sensor data related to the container components in order to perform container inspection tasks. However, it is also conceivable to set up several sensor devices for this purpose, which collect sensor data related to the container components from several shooting directions and / or collect data on container components transported on different tracks in a multi-track transportation area.

[0024] The sensor data is preferably spatially resolved sensor data, which specifically reflects the characteristics to be acquired (e.g., color values ​​and / or grayscale values ​​and / or brightness values) of a certain area of ​​the container component. Preferably, the spatially resolved sensor data indicates (or is able to indicate) a sensor data curve that depends on at least one spatial and / or geometric coordinate, and preferably depends on at least two spatial and / or geometric coordinates.

[0025] Sensor data, such as sensor data captured by a camera, can be, for example, color values ​​and / or grayscale values ​​and / or brightness values.

[0026] Sensor data acquired by a LIDAR device can be RGB values ​​and / or intensity values, which are acquired and stored for each data point together with or depending on its X, Y, and Z position values.

[0027] It is also conceivable that the sensor data is frequency-resolved sensor data.

[0028] Therefore, for example, for each sensor data point, at least one intensity value can be acquired that depends on the frequency of the radiation acquired by the sensor device, so that the sensor data indicates a sensor value curve that depends on the frequency.

[0029] It is also conceivable that the sensor data is spectrometer sensor data, preferably generated by a mass spectrometer, which is used, for example, as an odor sensor. It is conceivable that this is used to analyze, for example, the composition of gases and / or air and / or air mixtures (e.g., a gaseous environment in a container). Here, the acquired sensor data can, for example, indicate an intensity curve that depends on the mass-to-charge ratio of the atoms and molecules contained in the gaseous environment.

[0030] Preferably, the acquisition of sensor data, particularly spatially resolved sensor data, is optical. More preferably, the spatially resolved sensor data is a camera image.

[0031] Preferably, sensor data related to the container component to be inspected is acquired while these individual container components are being transported at a constant, and in particular, unreduced, transport speed—that is, while the container components are in motion. In other words, the container components are not slowed down and / or stopped in order to acquire sensor data. This provides the advantage of high throughput and production speed for the container handling system.

[0032] For optical acquisition of sensor data, a container component can be provided, preferably with illumination of the container, such as incident light illumination and / or transmitted light illumination.

[0033] Real-time evaluation devices, particularly processor-based real-time evaluation devices (for performing at least one container inspection task), evaluate using container inspection models, particularly real-time evaluation of sensor data, especially spatially resolved sensor data (within the framework of computer-implemented method steps).

[0034] The container inspection model is preferably a machine learning container inspection model, which includes a set of parameters that are set to values ​​learned as a result of a machine learning method (i.e., a training process).

[0035] The real-time evaluation device is preferably part of a container handling system and / or (in more detail below) a container inspection apparatus.

[0036] According to the present invention, (preferably a real-time evaluation device) is pre-defined and / or pre-defined a set of container component features based on a machine learning method, preferably based on (the above) mentioned above.

[0037] Preferably, based on a predetermined set of container component characteristics, the acquired, particularly spatially resolved, sensor data is evaluated (preferably by a real-time evaluation device) for a system inspection task different from the container inspection task (or for performing a system inspection task different from the container inspection task). Preferably, depending on the inspection results of the performed system inspection task, at least one system inspection parameter is determined, which is provided for (at least partially automatically, preferably fully automatically) controlling and / or regulating the container handling system. Here, it may be that, in order to perform the system inspection task, the acquired sensor data has specifically been selected, which has already achieved a predetermined inspection result when performing the container inspection task (e.g., no predetermined defects were identified or the sensor data was generally classified as defect-free).

[0038] Additionally or alternatively, the acquired, particularly spatially resolved, sensor data is preferably evaluated (particularly for system inspection tasks different from container inspection tasks or for performing system inspection tasks different from container inspection tasks) with respect to predetermined and / or predeterminable reference data, by determining a similarity parameter characterizing the similarity between the sensor data and reference data. Here, the similarity parameter, or parameters derived therefrom, are provided for controlling and / or regulating the container handling system. Here, the similarity parameter is preferably determined based on a predetermined set of container component features, particularly preferably based on machine learning methods (as described above). Here, the similarity parameter can be used as a system inspection parameter. It is also conceivable that the system inspection parameter is determined depending on the similarity parameter, preferably depending on multiple similarity parameters.

[0039] Preferably, the container handling system is controlled and / or regulated (at least partially automatically, preferably fully automatically) based on at least one (determined) system inspection parameter and / or based on the inspection results of the performed system inspection task (and / or based on similarity parameters). Preferably, the execution of at least one processing step performed by the container handling system, particularly preferably at least one processing step (to be performed on multiple container components), and / or the operating state of the container handling equipment are controlled and / or regulated (based on at least one system inspection parameter and / or based on the inspection results of the performed system inspection task and / or based on similarity parameters).

[0040] In other words, the first preferred embodiment of the method proposes not only to generate a container inspection model in a machine learning method to evaluate sensor data, in which the sensor data can be classified, for example, according to defects and / or types of container components identified therein, but also to provide means, in the form of a set of container component features obtained within the framework of a preferred machine learning method, by means of which the sensor data can also be evaluated with respect to the function and control / regulation of the container processing system.

[0041] In other words, the second preferred embodiment of the method proposes to evaluate the sensor data not only by means of a machine learning container inspection model for performing the container inspection task (e.g., for classifying defects and / or types of container parts identified in the sensor data), but also by means of similarity analysis with predetermined reference data.

[0042] Preferably, the reference data is sensor data acquired by a sensor device, particularly spatially resolved (related to container components), which is used as a reference, i.e., reference sensor data (hereinafter referred to as reference sensor data). The sensor device can be, in this context, a sensor device that acquires sensor data, particularly spatially resolved sensor data (requiring the determination of storage parameters regarding the sensor data).

[0043] Additionally or alternatively, the sensor device may be a sensor device (e.g., structurally identical) to the sensor device that acquires sensor data, particularly spatially resolved sensor data (of the container handling system). For example, the sensor device that acquires reference sensor data may be another sensor device within the same container handling system.

[0044] In addition, sensor data acquired by structurally identical sensor devices from different container handling systems can be used as reference sensor data, for example. This provides the advantage of being able to check, for example, whether very rare defects or features arising from faults in different container handling systems that occur in the container components being handled also occur in the container handling system under consideration.

[0045] The reference data is preferably stored in the storage device of the container handling system, and particularly preferably in the container inspection device that includes the sensor device.

[0046] Specifically, the collected, especially spatially resolved, sensor data is compared with reference data, and similarity parameters are determined based on the comparison results.

[0047] Here, the first and second preferred embodiments of the method can exist independently as a single method. Alternatively, both the first and second preferred embodiments of the method can be implemented within a common method. Therefore, the features listed below can relate to the first and second preferred embodiments described above, or to the implementation of either of the two preferred embodiments of the method.

[0048] Preferably, the similarity parameter is a non-discrete parameter, which in particular cannot take only two values ​​or a finite (fixed) number of values. Preferably, the similarity parameter is a continuous parameter. Preferably, the similarity parameter represents the degree of similarity.

[0049] Preferably, the container detection model assigns at least one of a plurality of predetermined categories to the collected sensor data, thereby enabling a more refined evaluation of the sensor data by determining non-discrete similarity parameters. This allows for the identification, for example, at a very early stage, of slowly evolving trends (which may arise, for example, during component aging or changes in production materials in the container handling system and affect the handling of container components), before the changing handling of container components affects their classification.

[0050] In a preferred method, the set of container component features is a set of container component features that are preferably automatically obtained within the framework of a machine learning method, particularly extracted by a neural network. Preferably, the set of container component features comes from a machine learning method (automatically).

[0051] Here, the machine learning method is preferably a machine learning method performed to obtain a container inspection model. However, it is also conceivable that the set of container component features is obtained from a machine learning method trained on a different container inspection task (which needs to be performed by a real-time evaluation device based on the collected, particularly spatially resolved, sensor data).

[0052] Here, the training container inspection task is preferably different from the container inspection task. Preferably, container component features are extracted during or in the course of a machine learning method performed with respect to the training container inspection task. By executing this machine learning method, a (trained) algorithm or (trained) (machine learning) model should be obtained for performing or executing the training container inspection task.

[0053] Preferably, the set of extracted container component features is a set of container component features that are (automatically) extracted within the framework of the machine learning method performed on the training container inspection task.

[0054] The fact that a container inspection task can be trained using a method different from the container inspection task itself has the advantage that a single training process or machine learning method (i.e., a training process or machine learning process performed within the framework of training the container inspection task) is sufficient for feature extraction. The extracted container component features obtained here are then used for other container inspection tasks.

[0055] For example, the feature vector of the bottle extracted from the (camera) image can be used in several inspection tasks / classifications.

[0056] Two examples of inspection tasks / categories:

[0057] 1. Does the feature vector show the brown bottle?

[0058] 2. Does the feature vector show sealed bottles?

[0059] Both can be identified in the image, and therefore can also be identified in the extracted feature vector.

[0060] In the first case, the feature vectors of brown and other colored bottles are used to "train" a machine learning algorithm (such as kNN). In the second case, the feature vectors of sealed and unsealed bottles are used.

[0061] These features only need to be extracted once per image.

[0062] In another preferred method, the learning method for extracting the set of features of container components is a supervised learning method. Preferably, a set of training data is used to perform the supervised learning method, which includes (collected) sensor data related to the container components, which is labeled or marked with the inspection results (e.g., multiple predetermined categories) of a predetermined (training) container inspection task to be obtained in the corresponding situation.

[0063] In another preferred approach, the supervised learning method is the K-Nearest Neighbors algorithm (also abbreviated as "k-NN" or "KNN"). Advantageously, this is a simple algorithm that can easily adapt to newly added training patterns. The K-Nearest Neighbors algorithm only requires a K value and a distance metric (migration metric), which is far fewer values ​​required compared to other machine learning algorithms.

[0064] In another preferred approach, conventional machine learning algorithms are used (as a learning method), such as decision tree learning (where decision trees are nonparametric supervised learning algorithms, preferably having a hierarchical tree structure, such as for classification and regression tasks), random forests or random decision forests, logistic regression, k-means clustering, and support vector machines (abbreviated as SVM).

[0065] Here, the extracted container component features are not, in particular, predetermined features, nor are they selections of predetermined features (e.g., user-defined features). The extracted container component features are, in particular, abstract features that represent or characterize, for example, contrast, parameters characterizing the frequency of straight lines (e.g., the number of straight lines), brightness or brightness curves, the shape of contours (lines) and / or boundary lines, angles, shapes, number of angles, curvature, combinations thereof, etc.

[0066] Preferably, (in particular, all) container component features are automatically generated, preferably automatically generated within the framework of machine learning methods (and in particular not selected).

[0067] Preferably, the set of extracted container component features is not adjusted and / or changed, even when defining (new and / or further and / or adjusted) container inspection tasks and / or new / modified system inspection tasks and / or prescribing new / modified reference data or reference sensor data.

[0068] Preferably, the set of container component features can be accessed, and in particular, individual container component features within that set can also be accessed. That is, the set of container component features is not inherently implicit in the image evaluation algorithm (similar to a "black box"), but is stored in such a way that they can be accessed independently and separately. Particularly preferably, the set of container component features can also be replaced individually (particularly independently of other software modules). It is also conceivable that the set of container component features can be output and / or transmitted solely for its own sake.

[0069] It is conceivable that, for example, the user of the container processing system predetermines the number of container component features in the set of container component features to be extracted or to be extracted, and / or the number of container component features in the set of container component features to be extracted or to be extracted is, for example, predetermined by the user of the container processing system. Preferably, when determining the set of container component features to be extracted, a predetermined number of container component features is considered. For example, the user of the container processing system may transmit this predetermined number of container component features to an external server, which determines the set of container component features to be extracted. However, it is also conceivable that the number of container component features is predetermined by the manufacturer of the container processing system, and is particularly not influenced by the container processing system (or its operator).

[0070] In another preferred method, the feature space is stretched by a set of extracted container component features, or a feature space stretched by a set of extracted container component features is provided. In other words, a feature space can be formed by a provided set of extracted container component features / the feature space is stretched by a provided set of extracted container component features.

[0071] Preferably, a distance metric (spread metric) related to the feature space is provided.

[0072] Preferably, a similarity parameter is determined using a distance metric (spacing metric). Here, sensor data, particularly spatially resolved sensor data, and / or reference sensor data are preferably represented in a feature space (as feature vectors). The spacing between these two feature vectors is preferably determined using a distance metric (spacing metric). This spacing, or a parameter characterizing this spacing, is preferably used as a similarity parameter between the acquired, particularly spatially resolved sensor data, and the reference sensor data.

[0073] The reference data may, for example, already be a representation of (reference) sensor data in the feature space. That is, the reference data may, for example, already be a feature vector (or a representation of that feature vector). Reference data that has already been used as a (pre-determined) feature vector indicator has the advantage of significantly smaller data volume.

[0074] In other words, a distance metric (spacing metric) associated with the feature space is preferably provided, wherein the distance metric (spacing metric) is used additionally or alternatively (preferably a real-time evaluation device) as a similarity metric between sensor data of different container components, preferably different containers, particularly spatially resolved, and / or as a similarity metric between the acquired sensor data and reference sensor data.

[0075] Preferably, (within the framework of performing system inspection tasks) the similarity between the collected, particularly spatially resolved, sensor data of one container component and sensor data of another container component (e.g., intended as reference sensor data) is assessed using distance metrics.

[0076] Preferably, based on the extracted set of container component features, a feature vector is created for all acquired (particularly spatially resolved) sensor data, i.e., for example, for each captured camera image. The feature vector can be, for example, a 256-dimensional vector.

[0077] Preferably, the feature vector has a maximum of 512 dimensions, more preferably a maximum of 256 dimensions, and particularly preferably a maximum of 128 dimensions. More preferably, the feature vector has a minimum of 16 dimensions, more preferably a minimum of 32 dimensions, more preferably a minimum of 64 dimensions, and particularly preferably a minimum of 128 dimensions. However, in principle, it is also possible to use feature vectors with more than 512 dimensions or feature spaces corresponding to higher dimensions.

[0078] Preferably, the distance between the feature vector of sensor data collected about the first container component and the feature vector of sensor data collected about the other container component can be determined by means of a distance metric.

[0079] In other words, distance metrics are preferably used as similarity metrics to assess the similarity between acquired, particularly spatially resolved, sensor data and reference data, particularly reference sensor data, in order to determine similarity parameters.

[0080] In another preferred approach, Euclidean metric and / or cosine similarity in the feature space are used as distance metrics. Cosine similarity (also known as "cosine similarity" or "cosine distance") is a measure of the similarity between two vectors, where the cosine value of the angle between the two vectors is determined. Cosine similarity can be specifically understood as the degree of directional consistency between two vectors. Specifically, the cosine similarity between two vectors a and b can be calculated by dividing the standard dot product of vectors a and b by the Euclidean norm of a and the Euclidean norm of b, i.e.: cosine similarity = (a·b) / (||a||||b||).

[0081] A relatively small distance between two feature vectors (in the feature space) obtained using a distance metric is preferably considered as low similarity between two sensor data points (or sensor data and reference (sensor) data) corresponding to the respective feature vectors. Conversely, a relatively large distance between two feature vectors (in the feature space) obtained using a distance metric is preferably considered as high similarity between two sensor data points (or sensor data and reference (sensor) data) corresponding to the respective feature vectors.

[0082] In another preferred method, the movement of the container component through the container handling system is tracked in at least segmented stages, based on an evaluation of sensor data collected about the container component according to a set of features of the container component. This provides the advantage of being able to determine whether the container component has arrived at / passed through the sensor devices or several sensor devices of the container handling system, and to check whether the container component being checked has repeatedly arrived at the sensor devices multiple times.

[0083] In another preferred method, container components are tracked without any processing steps performed or set for personalization. That is, for example, no unique code is set for personalization of the container component, such as a QR code. The applicant has found that container components can be identified and re-identified solely based on different representations of container component features (extracted in machine learning methods).

[0084] That is, for example, a sensor device (or real-time assessment device) can identify the contamination status of a contaminated container and, in response to an inspection task (contamination identification in this case), send the container to a cleaning unit for cleaning. After cleaning, the container can be transported back to the container inspection unit, particularly the sensor device, using transport equipment. Based on an assessment of the sensor data collected about the container after a cleaning cycle, it is advantageous not only to identify that the container (still) is contaminated, but also to identify that the container (with this contamination) has been inspected by the sensor device once and has therefore undergone (at least one) cleaning process. This can, for example, prevent contamination that cannot be removed by the cleaning equipment from causing the same container to undergo or have already undergone a cleaning-inspection-cycle, thus consuming processing capacity for other containers.

[0085] Preferably, the number of times a container component has been detected by the sensor device or has arrived at the sensor device is counted (particularly through a real-time evaluation device and / or container inspection apparatus). Preferably, the maximum number of times (by the operator) can be predetermined and / or stored in the storage device of the container inspection apparatus and / or container handling system, and the container handling system, particularly the rejection device for rejecting container components inspected by the sensor device, is controlled and / or adjusted based on the comparison between the predetermined maximum number and the corresponding number of times the container component has arrived at the sensor device and / or been detected by the sensor device.

[0086] In another preferred method, the sensor data collected about the container component is evaluated based on a set of features of the container component in such a way that at least one identification parameter characterizing the container component is determined.

[0087] Preferably, a feature vector is used as the discrimination parameter, by which the associated acquired sensor data can be represented or presented in the feature space. Preferably, this discrimination parameter is stored on a storage device (particularly the storage device of a container inspection device and / or a real-time evaluation device and / or a container handling system). This provides the advantage that it eliminates the need to store sensor data, thereby saving storage space.

[0088] Preferably, the discrimination parameter is compared with (at least one) discrimination parameter determined for other sensor data acquired by the sensor device. Preferably, the distance between the two feature vectors used as discrimination parameters is determined. Preferably, based on the distance determined (using a distance metric), it is determined whether the respective associated detected sensor data are sensor data of the same container component.

[0089] Preferably, a threshold spacing is predetermined and / or (by the operator) can be predetermined for this purpose. When the distance (spacing) between two feature vectors (or discrimination parameters) (determined by means of a distance metric) is less than the threshold spacing, it is determined that the associated detected sensor data belongs to the same container component.

[0090] In another preferred method, based on an evaluation of the sensor data collected regarding the container components, and using a set of features of the container components, it is checked whether individual container components have (preferably repeatedly) arrived at the sensor devices of the container handling system or at least another sensor device. Preferably, a similarity parameter characterizing the similarity of the respective collected sensor data is determined for this purpose. The similarity parameter can be determined, for example, using the aforementioned distance metric (as a similarity metric).

[0091] Here, at least one other sensor device is positioned downstream, for example, regarding the transport direction of the container component. Preferably, in addition to the sensor device, this at least one other sensor device also collects sensor data associated with (preferably each) container component. Preferably, the sensor device collects data about the container component from the same acquisition direction associated with it. Preferably, whether it is the same container component can be determined by using a similarity parameter (as presented above) based on a distance metric according to the respective collected sensor data. This allows determination of whether and / or when the container component arrives at the at least one other sensor device. If multiple additional sensor devices are provided (e.g., along the entire transport path through the container handling system), the movement of (individual) container components through the (entire) container handling system can be tracked.

[0092] In another preferred method, to check for repeated arrivals of the sensor device acquiring data from the individual container component, a discrimination parameter (e.g., a feature vector) characterizing the container component is determined based on a set of features of the container component, and the similarity to sensor data subsequently acquired by the sensor device is determined based on the determined discrimination parameter. For example, similarity can be determined by measuring the distance between the discrimination parameter selected as the feature vector and the feature vector determined with respect to the sensor data subsequently acquired by the sensor device.

[0093] In another preferred method, when it is determined that a container component has repeatedly arrived at the sensor device, the container component is preferably prevented from being transported to the sensor device again by individually modifying at least one previous processing step associated with the container component, particularly preferably removing the container component from the container component stream. That is, for example, the cleaning steps that the container component needs to undergo can be modified, and / or the container component can be (permanently) removed from the container component stream.

[0094] In another preferred method (not only in the second preferred embodiment described above, but also in the first preferred embodiment of the proposed method), reference data is predetermined and / or can be predetermined, and control and / or adjustment is made based on a determined similarity parameter that characterizes the similarity between the reference data and sensor data acquired (with respect to at least one container component, preferably with respect to each of the multiple container components).

[0095] In another preferred method, at least one processing step and / or at least one operating state of the container handling system is modified and / or adjusted based on the inspection results of the system inspection task. Preferably, the processing step and / or the at least one operating state is controlled and / or adjusted based on the inspection results of the system inspection task, and particularly preferably based on similarity parameters determined according to the collected sensor data and (in particular) predetermined reference (sensor) data.

[0096] For example, sensor data (as reference sensor data) or feature vectors for container components (preferably using a container inspection machine learning model) are pre-defined as reference data, and these container components are classified / belong to a classification such that the associated container components are not defect-free and / or need to be rejected. That is, container components mapped in the reference sensor data or associated with the feature vectors may have predetermined defects, which are identified by a real-time evaluation device (e.g., using a machine learning container inspection model) when performing a container inspection task.

[0097] Using these reference data, additional sensor data acquired by the sensor device (which has been identified as defect-free during container inspection / is identified as defect-free (at least regarding predetermined defects)) is preferably investigated to determine whether it (still) has (e.g., a minimum similarity predetermined by the operator) to these (defective) reference data and / or what degree of similarity it has with these reference data. As described above, the similarity or degree of similarity between the corresponding sensor data and the reference data can be determined. Specifically, the distance between the representation of the sensor data in the feature space and the representation of the reference data in the feature space (in each case, a feature vector) is calculated using a distance metric.

[0098] If the calculated gap is lower than a predetermined value (which indicates minimum similarity), the associated container component is considered sufficiently similar to the reference data (regarding the predetermined minimum similarity). In this case, for example, it could be a container component that exhibits slight manifestations of a predetermined defect but (still) meets (the necessary, e.g., predetermined by the operator) quality requirements.

[0099] For example, recyclable glass bottles (as container components) may have so-called "scuffing" on their outer glass surface. This scratching, or abrasion of the glass surface, can occur, for example, during repeated passage through a recyclable glass bottle filling system. Minor scratches are visually identifiable but do not cause a quality reduction. Only when scratches are very noticeable will the glass container be removed from the container stream and sent, for example, to a recycling system.

[0100] By performing a container inspection task, container components (here, glass bottles) that exhibit very obvious “scratches” need to be removed from the container stream and sent to the recycling system are identified. It is now proposed that performing the system inspection task by additionally comparing sensor data collected for each (all) inspected container component with reference data provides information about the presence and number of container components (here, glass bottles) in the container component stream inspected by the sensor devices that have not been rejected and / or do not need to be rejected, exhibiting scratches of a predetermined degree. For example, this can indicate information about multiple container components processed in the container handling system, such as their age and / or condition.

[0101] More preferably, the container handling system is configured, for example, to control the transport speed and / or modify the guiding device used to guide the containers to be transported, based on the scratches that are identified as having smaller performance, i.e., (typically) based on similarity parameters determined from multiple collected sensor data that are preferably associated with predetermined reference data.

[0102] As an example of a defect, the one described above is the so-called "scratches".

[0103] Additionally or alternatively, it is preferable to predetermine reference data that maps to container components with predetermined defects, which may occur during at least one processing step of the container component. In this way, it is advantageous to draw conclusions about malfunctions and / or harmful developments in the processing flow of the container handling system, for example, during blow molding.

[0104] Here, the cause of the predetermined defect may lie in the execution of the processing steps themselves and / or the state of the container handling system (or container handling equipment), such as the existing process temperature, such as the temperature in the plastic preform heating equipment.

[0105] Such reference data or reference sensor data may come from previous operations of the container handling system or other (especially structurally similar) container handling systems.

[0106] Preferably, (as described above) (within the framework of the system check task execution) the sensor data collected by the sensor device is evaluated with respect to predetermined reference data, specifically by determining (e.g., as explained above) a similarity parameter (preferably by using a distance metric to determine the spacing).

[0107] Preferably, the system inspection task is performed on multiple sensor data collected (regarding container components), wherein no predetermined defect is identified during the container inspection task, and / or the container component corresponding to the defect is not rejected from the container component stream according to the inspection results of the container inspection task. This provides the advantage of more detailed analysis of the sensor data that are judged to be of acceptable quality. In this way, even slightly apparent design flaws in container components can be detected.

[0108] Preferably, at least one system inspection parameter is determined based on a system inspection task performed on multiple collected sensor data and / or based on similarity parameters determined for each of the multiple collected sensor data.

[0109] Here, the system inspection parameters identified characterize the deviation of at least one operating state of the container handling system and / or the handling equipment / a handling equipment from the normal and / or ideal state.

[0110] Additionally or alternatively, the determined system inspection parameters characterize the aging of at least one component of the container handling system and / or a handling device and / or the fault conditions present in the container handling system and / or a handling device.

[0111] This provides the advantage that errors or aging (such as wear) or faulty conditions that may exist in a container handling system can be identified before they actually manifest as defects in container components (identified by real-time evaluation equipment or with the help of container inspection models).

[0112] Therefore, depending on the determined system inspection parameters, appropriate countermeasures and / or controls / regulations can be implemented for at least one process in the container processing system, which can counteract what is considered an unfavorable development observed in the container processing system. Preferably, so-called "predictive maintenance" can be performed depending on the system inspection parameters and / or depending on the determined similarity parameters.

[0113] It is also conceivable that, depending on the determined system inspection parameters and / or the determined similarity parameters, operators may be prompted to shorten maintenance intervals and / or perform maintenance.

[0114] Here, it is preferable to evaluate the similarity parameters determined for each of the multiple sensor data collected with respect to their relative frequency of occurrence in the container component flow (when determining system inspection parameters).

[0115] It is conceivable that the system check parameters are statistical parameters, in which, for example, similarity parameters of sensor data related to more than 1,000 container components, preferably more than 10,000, preferably at least 50,000, and particularly preferably at least 100,000 container components are taken into account or have been taken into account. This advantageously enables the control / regulation of the container handling system based solely on statistically significant system check variables.

[0116] Preferably, a certain amount of labeled and / or classified sensor data (e.g., images) for each application is used for training the network, ranging from 1,000 to 100,000 (e.g., 10,000 labeled and / or classified images per application).

[0117] Preferably, patterns are investigated using (automatic or AI-based) pattern recognition methods on similarity parameters determined with respect to predetermined reference data, such as the patterns arising from the corresponding spacing between the feature vectors of the respective acquired sensor data and the feature vectors of the reference data. In this way, for example, frequently occurring (especially unknown) features in the acquired sensor data can be identified, thereby identifying features related to the container component. Preferably, the identified features are responded to by, for example, storing and evaluating processing data and / or status data of the processing equipment for the corresponding container component present at the corresponding processing time point. Preferably, the evaluation results are taken into account when controlling and / or regulating the container processing system.

[0118] It is also conceivable that multiple clusters are formed for the feature vectors determined for multiple collected sensor data using cluster analysis (e.g., k-means algorithm), and the determined feature vectors are assigned to these clusters / the determined feature vectors are assigned to these clusters and / or the determined feature vectors can be assigned to these clusters. It is conceivable that the resulting clusters are evaluated in terms of known defects and / or features of the container component to be inspected. That is, the resulting clusters may, for example, map defect-free container components. Other resulting clusters may represent grouped sensor data, each mapping to a container component with the (same) known defects.

[0119] Furthermore, clusters can be obtained that group the collected sensor data, excluding container components with known defects. Data characterizing such clusters, such as centroids or cluster centers (and / or variances), can be output to the operator and / or used to evaluate any previously unknown defects and / or characteristics / attributes of the inspected container components.

[0120] It is also conceivable to use the centroids or cluster centers obtained through cluster analysis (especially in the feature space) as reference data.

[0121] In a further preferred method, at least one container inspection task is modified and / or adjusted and / or supplemented depending on the inspection results of the system inspection task. In this way, for example, the threshold for identifying defects in container components can be fine-tuned and / or newly identified defect types can be added.

[0122] In another preferred approach, a similarity parameter characterizes the similarity between acquired, particularly spatially resolved, sensor data and a predetermined and / or predeterminable plurality of, particularly spatially resolved, reference sensor data. That is, a similarity parameter representing or characterizing the similarity to several reference sensor data can also be considered.

[0123] In another preferred method, the operator of the container handling system, preferably using the human-machine interface of the container handling system, predetermines reference data, preferably spatially resolved reference sensor data and / or multiple, particularly spatially resolved reference sensor data. This provides the advantage that the operator can select, for example, the collected sensor data suggested for the operator as reference sensor data using an input device (e.g., designed as a touchscreen) of the container handling system (e.g., by means of an inspection device and / or the container handling equipment, particularly by means of a display device).

[0124] However, it is also conceivable that operators could transmit such reference data via the human-machine interface of the container inspection device and / or container handling system, with regard to which similarity parameters need to be determined. In this way, reference data can be determined in a user-friendly manner.

[0125] In another preferred method, the container inspection task is a classification task selected from a set of classification tasks, including: classifying container components and preferred containers as defective or non-defective (preferably binary classification) (good / defective); identifying and / or classifying the defect types of container components and preferred containers; identifying and / or classifying different types of container components and preferred containers (e.g., about ten different bottle types); identifying and / or classifying the outline and / or color of container components and preferred containers; and classifying the defects at the inspected container components (especially by means of the first...). Identifying and / or classifying error-free / error-prone execution of at least one processing step in a processing device; identifying relatively rare container component types and / or defect types in a container component flow (“rare” specifically means occurring at a frequency of less than 1 / 1000); label control; liquid level control; inspecting for bubbling behavior of liquids within containers; identifying container cracks and / or bottle neck breaks and / or container bottom area ruptures; identifying foreign particles attached to or present in or within container components and / or containers; and similar classification tasks and combinations thereof.

[0126] The machine learning container inspection model is preferably based on an (artificial) neural network. Preferably, the neural network is formed as a deep neural network (DNN), wherein the parameterizable processing chain has multiple processing layers, and / or so-called convolutional neural networks (CNNs) and / or recurrent neural networks (RNNs) and / or other DNN layer classes.

[0127] Preferably, the machine learning container inspection model used by the real-time evaluation device is a container inspection model that has already been learned or (completed) trained. In other words, the container inspection model, which is to be trained using the training dataset to be generated or to undergo further training for greater accuracy, is in a state where the training process has been completed. It is conceivable that the container inspection model has been trained using a general training dataset, preferably independent of a specific or particular container processing system, or that the container inspection model is in a state where it has been trained using the aforementioned model.

[0128] Preferably, the (data to be processed), particularly spatially resolved sensor data (or data derived therefrom), is fed as an input variable to the container inspection model or (artificial) neural network. Preferably, the container recognition model or artificial neural network maps the input variable to the output variable based on a parameterizable processing chain, wherein preferably a container inspection category is selected as the output variable or preferably multiple container inspection categories are selected as the output variable.

[0129] Preferably, the multiple (different from each other) container inspection categories refer to different defect types of container components. For example, the different container inspection categories may refer to different areas of the container component, such as selecting a group that includes the bottom area, outer wall area, side wall area, corrugations within the bottom area, support ring area, mouth area, and similar areas, as well as combinations thereof.

[0130] Different container inspection categories can also be derived from different container inspection tasks, such as bottom inspection of container components, sealing surface inspection, side wall inspection, thread inspection, sealing inspection, liquid level inspection, label inspection, tamper protection inspection, suspended solids inspection, and residual liquid inspection.

[0131] Additionally or alternatively, container inspection categories may also refer to different defect types (e.g., cracks, fissures, breaks, etc.) and / or failure types (exactly one or at least one of them), which may be selected from a group including impurities, breakage, cracks, different crack types, (glass) fragments, spalling, contamination spots, dirt types, material distribution associated with defects, and similar defects and combinations thereof.

[0132] Preferably, a machine learning container inspection model or artificial neural network is trained using predetermined training data, wherein the parameterizable processing chain is parameterized through training.

[0133] In a preferred approach, training data is used during the training of the container inspection model. This training data includes multiple spatially resolved sensor data (of the container) collected by at least one sensor device. This provides the advantage that the training process has been specifically coordinated for the container inspection apparatus to be configured, thereby allowing for direct consideration, for example, of the specific conditions of that particular container inspection apparatus, such as the optical characteristics of the sensor devices or specific lighting conditions within the container inspection apparatus.

[0134] Preferably, the spatially resolved sensor data (acquired by at least one sensor device) used as training data includes (container) type features and / or classification features. Preferably, the spatially resolved sensor data, along with their respective corresponding (container) type features and / or classification features, are stored as a training dataset (particularly on a non-volatile storage device and / or on that non-volatile storage device). Preferably, multiple training datasets are generated in this manner. The classification features can be used to classify the (as described above) containers. That is, the spatially resolved sensor data corresponding to the containers can be classified, for example, by the type of defects present therein.

[0135] The non-volatile storage device can be a storage device that is a (fixed) component of at least one sensor device and / or container inspection device. It is conceivable that this storage device is a ring memory in which the oldest sensor data is overwritten when the storage capacity is reached (therefore the stored sensor data is only available for a limited time).

[0136] Non-volatile storage devices can also be storage devices for container processing systems, which are designed, for example, as fixed memory.

[0137] "Non-volatile" can also mean that the selected image or sensor data and / or sensor data stored on a non-volatile storage device are only retained for a specific period of time, i.e., they can be deleted after a set time or after a parameterized operation (by an assembly worker). "Non-volatile" can also mean that the image or sensor data does not need to be retained or stored after the machine is turned off.

[0138] "Non-volatile" should also mean that the image data to be "held" and / or the sensor data to be stored must be available for parameterization in the simplest case.

[0139] It is also conceivable that, additionally or alternatively, "non-volatile storage device" is understood to mean that the image data to be "held" and / or the sensor data to be stored is still retained even if the storage device is not powered.

[0140] Training a network typically requires 10,000 or more labeled and / or classified images / sensor data per application. This labeling and / or classification work can be done locally or centrally by image processing experts.

[0141] Preferably, the working operation is a continuous (production) operation of the container inspection device and / or a continuous (production) operation of the container handling system (e.g., a container filling system) having the container inspection device. Specifically, the working operation can be a production operation. Specifically, the working operation is not a testing operation and / or maintenance operation and / or setup operation, in which the transport speed of the container or container component (as it passes through the container inspection device) is reduced compared to the transport speed in the working operation.

[0142] Preferably, the sensor device is selected from the group comprising: image capturing devices, such as cameras (preferably black-and-white and / or color cameras); CMOS sensors (CMOS is an abbreviation for Complementary metal-oxide-semiconductor); CCD sensors; 3D sensors; X-ray based image capturing devices; optical elements; thermal imaging cameras; stereo cameras; LiDAR cameras; odor sensors; and / or (mass) spectrometers; and similar devices and combinations thereof.

[0143] In a preferred embodiment, the transport device transports the container from a first processing device to another (or second) processing device (and / or the transport device is particularly suited to and designed to transport the container from the first processing device to another (or second) processing device).

[0144] Preferably, the first processing equipment and / or the other processing equipment is selected from a group comprising: injection molding apparatus for producing injection molded parts (e.g., preforms); cleaning apparatus for cleaning containers and / or container components; crushing equipment for crushing containers; filling apparatus for filling containers; molding apparatus, particularly blow molding machines, for forming plastic preforms into plastic containers; sealing equipment for sealing containers; labeling apparatus; marking equipment; sorting equipment; packaging equipment (for sealing and / or heat shrink packaging); equipment for bundling container components and / or containers; determining equipment for determining the composition of a substance or mixture of substances, such as determining the composition of the gaseous environment and / or air within or at a container component, preferably within or at a container, such as mass spectrometers and / or odor sensor devices; and similar equipment and combinations thereof.

[0145] Preferably, the container component flow is a sequential or sequential (particularly continuous) flow of container components (along the transport path). For example, the container component flow can be a flow of containers that are sequential or sequential (along the transport path). Here, the container component flow can be guided or transported in a single-channel or multi-channel manner (with the aid of transport equipment) in certain areas, preferably within the entire container inspection device (as a dense flow). Preferably, at least one sensor device is assigned to each channel of the container component flow, and each container component in the container component flow located in that channel is sampled.

[0146] The transport equipment may also be a high-density transport vehicle, used for transporting multiple container components, preferably containers, preferably in a multi-channel and / or unordered manner. The transport equipment may also be a buffer, used for buffering multiple container components, preferably containers, preferably in a multi-channel and / or unordered manner.

[0147] The container components, preferably the container, can preferably be transported or guided at least in sections and preferably along the entire transport area (by the transport equipment) upright or vertically.

[0148] Preferably, the transport equipment is suitable for and intended to guide or transport multiple container components, preferably containers, at least in segments, preferably along the entire transport area under back pressure.

[0149] Preferably, the transport equipment is adapted and designed (at least within the transport area) to transport and / or guide at least one container component (preferably at least one container) per hour, preferably at least 5,000 (especially those to be inspected) container components (preferably containers) per hour, preferably at least 20,000 (especially those to be inspected) container components (preferably containers) per hour, preferably at least 100,000 (especially those to be inspected) container components (preferably containers) per hour, preferably at least 140,000 (especially those to be inspected) container components (preferably containers) per hour, particularly preferably at least 180,000 (especially those to be inspected) container components (preferably containers) per hour, and this process is carried out within the operating operations of the handling equipment and / or container handling system. Preferably, the transport equipment is adapted and designed (at least within the transport area) to transport and / or guide up to 180,000 (especially preferably up to 200,000) (especially those to be inspected) container components (preferably containers) per hour, and this process is carried out within the operating operations of the handling equipment and / or container handling system and / or container inspection device.

[0150] Preferably, the transport equipment in a single-lane transport area is suitable and designed (at least within the single-lane transport area) to transport and / or guide at least 100,000 container components (preferably containers) per hour and / or at most 1,850,000 (particularly preferred, up to 200,000) container components (preferably containers) per hour, and the process is carried out within the working operation of the handling equipment and / or container handling system and / or container inspection device.

[0151] The container can be a preform, which is produced into a fully formed container through a molding process, and / or these preforms have so far been produced only through a primary molding step. The container can also be a fully formed container, which has reached its final shape, for example, through a molding process of a preform (e.g., an injection molded part or injection preform).

[0152] The container may be empty, still needing to be filled, and / or needing to be recycled and / or needing to be refilled. The container may be a filled container. Additionally or alternatively, the container may be a container sealed using (in particular exactly one) a container seal and / or capable of being sealed using (in particular exactly one) a container seal.

[0153] The container can be a single-use container or a reusable container.

[0154] The container is preferably (preferably sealed) and especially capable of being sealed, for containing liquids and / or flowable substances, such as pastes and / or ointments and / or gels, for example substances from the food industry, cosmetics industry or pharmaceutical industry.

[0155] It is also conceivable that the container is a container for holding liquids and / or solids, such as a container for holding contact lenses.

[0156] Preferably, the external storage device is a cloud-based (preferably non-volatile) storage device and / or an external server (containing the storage device), wherein the storage device is accessed, in particular, via the Internet (and / or via public and / or private networks, particularly at least partially wired and / or wirelessly connected). The external server should be specifically understood as an external server associated with the container processing system and / or real-time evaluation equipment, particularly a backend server.

[0157] External servers, for example, are backends, particularly for container processing system manufacturers or service providers, adapted to manage spatially resolved sensor data (particularly sensor data from multiple sensor devices and / or multiple container inspection devices), and / or execute machine learning methods related to the (training) container inspection task to be performed, and / or configure and tune real-time evaluation equipment. The functionality of the backend or external server can be executed on an (external) server farm. The (external) server can be a distributed system.

[0158] The present invention also relates to a control device for a container handling system for processing multiple container components for containers, and preferably for plastic containers and / or bottles.

[0159] Here, the container handling system has a transport device that transports multiple container components as a container component flow along a predetermined transport path from at least one handling device of the container handling system to at least another handling device of the container handling system.

[0160] Here, the container handling system has at least one sensor device for performing container inspection tasks, which is suitable for and designed to acquire, preferably optically, sensor data related to container components, particularly spatially resolved sensor data and preferably camera images.

[0161] Here, the container processing system has a real-time evaluation device that is suitable for and designed to evaluate sensor data, particularly spatially resolved sensor data, in real time using a machine learning container inspection model. This container inspection model includes a set of parameters that are set to values ​​learned as a result of the machine learning method.

[0162] According to the invention, the control device is adapted and designed to evaluate acquired, particularly spatially resolved, sensor data with respect to predetermined and / or predeterminable reference data by determining a similarity parameter characterizing the similarity between sensor data and reference data.

[0163] Here, the similarity parameter is preferably determined based on a predetermined set of container component features, particularly preferably based on machine learning methods.

[0164] According to the present invention, the control device is adapted and intended to provide similar parameters or parameters derived therefrom for controlling and / or regulating the container handling system, and / or to control and / or regulate the container handling system or the handling apparatus of the container handling system based on the similar parameters or parameters derived therefrom.

[0165] That is, within the framework of this invention, it is further proposed that not only should sensor data be evaluated using a container inspection model, but also that such sensor data be further evaluated based on the determination of the similarity between the collected sensor data and predetermined or predeterminable reference data. In this way, the AI-based evaluation model, which is considered a "black box," can be supplemented by similarity analysis of the mapped container components.

[0166] Preferably, the control device is adapted, suited, and / or intended to implement the above-described method and all method steps already described above in the context of the method, either individually or in combination thereof, particularly regarding the determination of similarity parameters and / or system check parameters. Conversely, the method can be configured with all the features described within the framework of the control device, either individually or in combination thereof.

[0167] The present invention also relates to a container handling system for processing multiple container components for a container. The container handling system includes a transport device for transporting the multiple container components as a container component flow along a predetermined transport path from or to at least one processing device of the container handling system, preferably from at least one processing device to at least another processing device.

[0168] Here, the container handling system has at least one sensor device for performing container inspection tasks, which is suitable for and designed to acquire, preferably optically, sensor data related to container components, particularly spatially resolved sensor data and preferably camera images.

[0169] Here, the container processing system also includes a real-time evaluation device suitable for and designed to evaluate sensor data, particularly spatially resolved sensor data, in real time using a machine learning container inspection model. This machine learning container inspection model includes a set of parameters set to values ​​learned as a result of a machine learning method.

[0170] According to the present invention, the container handling system has the above-described control device.

[0171] Preferably, the container handling system is adapted, suited, and / or designed to perform the methods described above for operating the container handling system, and to perform, individually or in combination, all the method steps already described in the context of the methods above. Furthermore, the container handling system and / or the handling equipment and / or the at least one other handling equipment and / or transport equipment may have or be equipped with at least one of the aforementioned features, individually or in combination with other features.

[0172] The present invention also relates to a computer method or computer method product, comprising method apparatus, particularly method code, which represents or encodes at least some and preferably all of the method steps of the method according to the invention, and preferably represents or encodes an embodiment of the preferred embodiments described, and is designed for execution by a processor device.

[0173] The present invention also relates to a data storage device that stores at least one embodiment or a preferred embodiment of a computer program according to the present invention.

[0174] This invention has been described with respect to containers or container components for containers. Here, the invention can also be transferred to injection-molded parts (injection preforms) or, more generally, to articles to be processed in a handling system (e.g., contact lenses to be manufactured and / or packaged), whose processing progress and / or characteristics and / or (defects / quality) status are controlled by at least one sensor device (for acquiring, in particular, spatially resolved sensor data, associated with each individual article to be inspected). The applicant reserves the right to claim rights to the subject matter involved.

[0175] The present invention also relates to a method for operating an article handling system to process a plurality of injection molded parts and / or articles, wherein the transport device transports the plurality of injection molded parts and / or articles as a parts flow along a predetermined transport path from at least one processing device of the article handling system to at least another processing device of the article handling system, wherein, in order to perform an article inspection task, at least one sensor device acquires, preferably optically, sensor data related to the injection molded parts and / or articles, particularly spatially resolved sensor data and preferably camera images, and a real-time evaluation device evaluates the sensor data, particularly the spatially resolved sensor data, in real time using a machine learning article inspection model, which includes a set of parameters set to values ​​learned as a result of a machine learning method.

[0176] According to the present invention, a set of article features is predetermined based on a machine learning method, preferably the machine learning method, and based on the predetermined set of article features, the collected, particularly spatially resolved, sensor data is evaluated for a system inspection task different from the article inspection task, wherein at least one system inspection parameter is determined depending on the inspection result of the performed system inspection task, and the at least one system inspection parameter is provided for controlling and / or adjusting the article handling system.

[0177] The present invention also relates to a control device for an article handling system for processing a plurality of injection molded parts and / or articles, wherein the article handling system has a transport device for transporting the plurality of injection molded parts and / or articles as a parts flow along a predetermined transport path from at least one processing device of the article handling system to at least another processing device of the article handling system, wherein the article handling system has at least one sensor device for performing an article inspection task, the at least one sensor device being adapted and intended to acquire, preferably optically, sensor data related to the injection molded parts and / or articles, particularly spatially resolved sensor data and preferably camera images, and wherein the article handling system has a real-time evaluation device adapted and intended to evaluate the sensor data, particularly the spatially resolved sensor data, in real time using a machine learning article inspection model, the article inspection model including a set of parameters set as values ​​learned as a result of a machine learning method.

[0178] According to the invention, the control device is adapted and intended to evaluate the acquired, particularly spatially resolved, sensor data with respect to predetermined and / or predeterminable reference data by determining a similarity parameter characterizing the similarity between sensor data and reference data, wherein the similarity parameter is preferably determined based on a predetermined, particularly preferred, set of article features based on machine learning methods, and wherein the control device is adapted and intended to provide the similarity parameter or a parameter derived therefrom for controlling and / or regulating the article handling system.

[0179] The other features described above within the framework of container components are applied in a similar manner to injection molded parts and / or articles.

[0180] For further advantages and implementation methods, please refer to the accompanying drawings:

[0181] in:

[0182] Figure 1 A schematic diagram of a container handling system according to a preferred embodiment of the invention is shown; and

[0183] Figure 2 A camera image is shown to illustrate the determined similarity parameters.

[0184] Figure 1 A schematic diagram of a container processing system 1 according to the invention for processing a container component 10 is shown, according to a first embodiment. The container component in this figure is a container 10 designed as a bottle.

[0185] Reference numeral 12 indicates the equipment arranged at container component 10 (the container in this figure). Figure 1 In the illustrated embodiment, an identification device is shown as an example of equipment, which is arranged on the bottle 10. For example, this is a (printed) QR code. Reference numeral 14 marks the container seal, which is another component of the container 10.

[0186] exist Figure 1 In the embodiment shown, a plastic preform is provided and fed by a transport device 6 into a heating device 20, where it is heated, and then expanded into a (plastic) bottle 10 in a blow molding apparatus as further processing equipment. The arrangement of the blow molding apparatus within the container processing system 1 is indicated by reference numeral 23. The bottle 10 may be equipped with an identification device 12, for example, by a personalization device (e.g., a printing device), thereby producing a bottle with the identification device 12.

[0187] The container component 9 can be transported from one processing unit to the next processing unit within the container handling system 1 by at least one transport device 6, and can also be transported within the processing unit. In this figure, the processing units shown are an inspection device 21, a filling device 22 for filling the bottles 10 with the product, a sealing device 24, a drying device 28, a labeling device 30, and a packaging device 32 for packaging the bottles 10 (in downstream order of the bottle transport direction).

[0188] Reference numeral 2 in each case indicates another container inspection device (e.g., arranged at the end of the production line and between the sealing device 24 and the drying device 28), which, for example, checks the filling height in the bottle and / or the correct arrangement of the seal on the bottle 10 and / or the safety ring and / or the correct labeling and / or packaging or other production data of the bottle 10.

[0189] In each case, reference numeral 4 indicates the sensor device, which in this figure is a camera, by means of which sensor data associated with each (inspected) container component 9 is collected, acquired, or recorded individually.

[0190] In each case, reference numeral 3 indicates a real-time evaluation device, by means of which sensor data collected by the sensor device 4 of the corresponding container inspection device 2 is evaluated to perform (predetermined and / or specified) container inspection tasks.

[0191] In a preferred proposed method, the detected sensor data is evaluated using a set of extracted features. The set of extracted features used is the result of (trained) feature extraction by a neural network pre-trained on a similar (inspection) task of image classification using extensive (training) data. However, in the final step of evaluating the extracted features, a classical classification method is also applied.

[0192] Reference numeral 50 denotes an internal server or internal storage device, and reference numeral 52 denotes an external server or external storage device, particularly a cloud-based storage device. For example, AI-based feature extraction can be performed on external server 52. Preferably, the set of extracted container component features can be stored on external and / or internal storage devices 50 / 52.

[0193] Reference numeral 5 in the attached figure indicates a storage device, which is a (fixed) component of the container inspection device 2 in this figure. Sensor data acquired by the sensor device 4 can be stored on the storage device 2.

[0194] Preferably, an image of a particular error or feature of a container component (as reference data) can be used as a reference. Using a preferred AI-based similarity metric, the similarity between the image captured by the camera and the reference image can be determined. This also allows for the identification of camera images showing containers exhibiting only slight variations of the searched specific error / feature, and containers associated with that error / feature being unidentified and / or rejected due to their slight variation. For example, if multiple containers exhibiting slight variations of the error are identified (gradually frequently) in camera images, it may indicate a continuously increasing deviation in a process parameter (such as the temperature of the cleaning fluid and / or the temperature in the heating device for heating the plastic preform).

[0195] Preferably, the container inspection device 2 or control device can determine a similarity parameter for this purpose, which characterizes the similarity between the acquired image or sensor data and predetermined and / or predeterminable reference data (e.g., reference images).

[0196] Depending on the respective similarity parameters determined, system inspection parameters can be determined, which characterize the (undesired) actual state (e.g., temperature state) of the container handling system and / or container handling equipment.

[0197] Depending on the parameters checked by the system, the state of, for example, the container processing system and / or container processing equipment can then be controlled / regulated to achieve the (desired) target state.

[0198] Figure 2 Twelve camera images are shown to illustrate the determined similarity parameters.

[0199] In particular, these (along with other camera images not shown) have been used to evaluate similarity. Figure 2 The results of sorting camera images based on their similarity (in descending order of similarity) are shown.

[0200] These camera images are captured within a frame that inspects the bottom of the container through its opening. The bottom is illuminated by a lighting device using transmitted light.

[0201] exist Figure 2 The first image marked with the reference numeral RSD in the upper left corner of the attached plane is used here as a reference image. Therefore, the distance between this image and itself, determined by means of a distance metric such as Euclidean distance, is 0.

[0202] Figure 2 The other camera images shown are sorted according to their respective distances from the reference image as determined by a distance metric (from left to right, then from top to bottom), thus exhibiting increasing distances, i.e. decreasing similarity.

[0203] The last camera image located in the lower right corner of the attached plane, and... Figure 2 The image shown has the largest spacing of 0.1848 compared to the other images, and is therefore the eleventh neighboring image of the reference image.

[0204] These camera images clearly demonstrate the high performance of the proposed method. Reference image RSD shows the bottom of a container with an embossed "BA". The camera images most similar to the proposed method, namely the first neighboring image ("Neighbour 1") and the second neighboring image ("Neighbour 2"), also show the same embossed "BA" (decreasing in sharpness). The third neighboring image ("Neighbour 3") shows a droplet in the center, its shape resembling the inner contour of the "B" in a circular pattern.

[0205] Figure 2 The proposed method, which evaluates similarity using a distance metric in a feature space (where each image is represented by a feature vector), allows all container bottoms with “BA” embossing to be sorted in their four nearest neighboring images.

[0206] This method offers the possibility of identifying an image similar to a predetermined reference image from the acquired (container flow) images. From this identification, conclusions can then be drawn regarding whether the container handling system is operating smoothly, or, on the other hand, whether intervention, control, and / or regulation should be implemented.

[0207] The applicant reserves the right to claim rights to all features essential to the invention disclosed in the application, whether individually or in combination, provided that such features are novel compared to the prior art. Furthermore, it should be noted that features that may be advantageous in themselves are also described in individual figures. Those skilled in the art will readily recognize that a particular feature described in a figure may be advantageous even without employing other features in that figure. Additionally, those skilled in the art will recognize that advantages can also be derived by combining several features shown in a single figure or different figures.

[0208] List of reference numerals

[0209] 1 Container handling system

[0210] 2.21 Container Inspection Device

[0211] 3. Real-time evaluation equipment

[0212] 4. Sensor devices

[0213] 6. Transportation equipment

[0214] 10 containers

[0215] 9 Container Components

[0216] 12. Equipment, Direct Printed Components

[0217] 14. Equipment, container sealing section

[0218] 20. Processing equipment, this section contains heating equipment.

[0219] 23. Processing equipment, this is the printing equipment.

[0220] 22. Processing equipment, this is the filling equipment.

[0221] 24. Processing equipment, this is the capping equipment.

[0222] 28. Processing equipment, this is the drying equipment.

[0223] 30. Processing equipment, this is the labeling equipment.

[0224] 32. Processing equipment, this section contains packaging equipment.

[0225] 50 Internal servers and storage devices

[0226] 52 External servers and storage devices

Claims

1. A method of operating a container processing system (1) for processing a plurality of container components (9) for a container (10), wherein a transport device (6) transports the plurality of container components (9) as a container component flow along a predetermined transport path from at least one processing device (20, 23, 22, 24, 28, 30) of the container processing system (1) to at least another processing device (23, 22, 24, 28, 30, 32) of the container processing system (1), wherein, in order to perform a container inspection task, at least one sensor device (4) acquires sensor data associated with the container components (9), and a real-time evaluation device (3) evaluates the sensor data in real time using a machine learning container inspection model, the container inspection model comprising a set of parameters set as values ​​learned as a result of a machine learning method. Its features are, A set of container component features based on machine learning methods is predetermined, and the collected sensor data is evaluated based on the predetermined set of container component features with respect to a system inspection task different from the container inspection task, wherein at least one system inspection parameter is determined depending on the inspection result of the performed system inspection task, the at least one system inspection parameter being provided for controlling and / or regulating the container processing system (1).

2. A method of operating a container processing system (1) for processing a plurality of container components (9) for a container (10), wherein a transport device (6) transports the plurality of container components (9) as a container component flow along a predetermined transport path from at least one processing device (20, 23, 22, 24, 28, 30) of the container processing system (1) to at least another processing device (23, 22, 24, 28, 30, 32) of the container processing system (1), wherein, in order to perform a container inspection task, at least one sensor device (4) acquires sensor data associated with the container components (9), and a real-time evaluation device (3) evaluates the sensor data in real time using a machine learning container inspection model, the container inspection model comprising a set of parameters set as values ​​learned as a result of a machine learning method. Its features are, The collected sensor data is evaluated with respect to predetermined and / or predeterminable reference data by determining a similarity parameter characterizing the similarity between the sensor data and the reference data, wherein the similarity parameter or a parameter derived therefrom is provided for controlling and / or regulating the container handling system (1).

3. The method according to claim 1 or 2, wherein, The set of container component features is the set of extracted container component features obtained within the framework of the machine learning method.

4. The method according to claim 3, characterized in that, The learning method for the set of features extracted from container components within its framework is a supervised learning method, preferably the K-nearest neighbor algorithm.

5. The method according to claim 1, characterized in that, Based on the evaluation of sensor data collected about the container component according to the set of characteristics of the container component, the movement of the container component by the container processing system (1) is tracked in at least segments.

6. The method according to claim 5, characterized in that, The tracking is performed on the container component without performing and / or performing any processing steps on the container component for personalization purposes.

7. The method according to claim 1, characterized in that, The set of features of the container component is used to evaluate the sensor data collected about the container component (9) in such a way that at least one identification parameter characterizing the container component (9) is determined.

8. The method according to claim 1, characterized in that, Based on the evaluation of the sensor data collected about the container components, and with the aid of the set of container component characteristics, it is checked whether an individual container component (9) has repeatedly arrived at the sensor device (4).

9. The method according to claim 8, characterized in that, In order to check the repeated arrival of the sensor device (4) that collects the individual container components (9), a discrimination parameter characterizing the container component (9) is determined based on the set of container component features, and the similarity with sensor data subsequently collected by the sensor device (4) is determined based on the determined discrimination parameter.

10. The method according to claim 9, characterized in that, When it is determined that a container component (9) repeatedly arrives at the sensor device (4), the container component (9) is prevented from being transported to the sensor device (4) again.

11. The method according to claim 1, characterized in that, Reference data is pre-defined or can be pre-defined, and the control and / or adjustment is based on a determined similarity parameter that characterizes the similarity between the reference data and sensor data acquired with respect to at least one container component (9).

12. The method according to claim 10, characterized in that, Depending on the inspection results of the system inspection task, at least one processing step and / or at least one operating state of the container processing system (1) may be changed and / or adjusted.

13. The method according to claim 12, characterized in that, Depending on the results of the system inspection task, at least one container inspection task may be modified and / or adjusted and / or supplemented.

14. A control device for a container handling system (1) for handling a plurality of container components (9) for a container (10), wherein the container handling system (1) has a transport device (6) for transporting the plurality of container components (9) as a container component flow along a predetermined transport path from at least one processing device (20, 23, 22, 24, 28, 30) of the container handling system (1) to at least another processing device (23, 22, 24, 28, 30, 32) of the container handling system (1), wherein the container handling system (1) has at least one sensor device (4) for performing a container inspection task, the at least one sensor device being adapted and intended to acquire, preferably optically acquire, sensor data associated with the container components (9), and having a real-time evaluation device (3) that evaluates the sensor data in real time using a machine learning container inspection model, the container inspection model including a set of parameters set to values ​​learned as a result of a machine learning method. Its features are, The control device is adapted and intended to evaluate the acquired sensor data with respect to predetermined and / or predeterminable reference data by determining a similarity parameter characterizing the similarity between the sensor data and the reference data, wherein the control device is adapted and intended to provide the similarity parameter or a parameter derived therefrom for controlling and / or regulating the container handling system (1).

15. A container processing system (1) for processing a plurality of container components (9) for a container (10), comprising a transport device (6) for transporting the plurality of container components (9) as a container component flow along a predetermined transport path from at least one processing device (20, 23, 22, 24, 28, 30) of the container processing system (1) to at least another processing device (23, 22, 24, 28, 30, 32) of the container processing system (1), wherein the container processing system (1) has at least one sensor device (4) for performing a container inspection task, the at least one sensor device being adapted and intended to acquire, preferably optically acquire, sensor data associated with the container components (9), and wherein the container processing system has a real-time evaluation device (3) that evaluates the sensor data in real time using a machine learning container inspection model, the container inspection model comprising a set of parameters set as values ​​learned as a result of a machine learning method, characterized in that, The container handling system (1) has a control device according to the preceding claim.

Images