US-12627884-B2 - Method for carrying out a setting operation of a container inspection apparatus and container inspection apparatus

US12627884B2US 12627884 B2US12627884 B2US 12627884B2US-12627884-B2

Abstract

A method for carrying out a setting mode of a container inspection apparatus including the following steps: a plurality of spatially resolved sensor data detected by the sensor device is stored on a non-volatile storage device, which is retrieved by a setting device in the setting mode; the setting device is provided with a set of test evaluation parameters, which are to be assessed in terms of a working mode of the container inspection device, in which the set of test evaluation parameters is set as a set of real-time evaluation parameters in the real-time evaluation device; for assessing the set of test evaluation parameters, the setting device determines at least one statistical assessment variable on the basis of the retrieved plurality of spatially resolved sensor data.

Inventors

Alexander Hewicker
Christof Will

Assignees

KRONES AG

Dates

Publication Date: 20260512
Application Date: 20221121
Priority Date: 20211215

Claims (20)

1 . A method for performing a setting mode of a container inspection apparatus, in which, in a working mode, a transport device transports containers to be inspected as a container stream along a predetermined transport path and at least one sensor device detects spatially resolved sensor data with respect to the containers to be inspected, in particular optically, and a real-time evaluation device evaluates the spatially resolved sensor data of the individual inspected containers in real time with the aid of an adjustable set of real-time evaluation parameters, comprising: for the setting mode, a plurality of the spatially resolved sensor data recorded by the sensor device is stored on a non-volatile storage device during the working mode, which is retrieved by a setting device during the setting mode; the setting device is provided with a set of test evaluation parameters to be evaluated with respect to a working mode of the container inspection apparatus, in which the set of test evaluation parameters is set in the real-time evaluation device as the set of real-time evaluation parameters; to evaluate the set of test evaluation parameters, the setting device determines at least one statistical evaluation variable on the basis of the retrieved plurality of spatially resolved sensor data.
2 . The method according to claim 1 , wherein the setting device determines the statistical evaluation variable independently of the set of real-time evaluation variables set in the real-time evaluation device.
3 . The method according to claim 2 , wherein the at least one statistical evaluation variable is determined based on the set of test evaluation variables while simultaneously inspecting containers based on the set of real-time evaluation variable in the working mode in the container inspection apparatus.
4 . The method according to claim 2 , wherein in the working mode an inspected container is rejected by a rejection device as a function of the evaluation carried out by the real-time evaluation device, and wherein the statistical evaluation variable is characteristic of a statistical rejection variable of the rejection device.
5 . The method according to claim 2 , wherein the statistical evaluation variable is determined locally separately from the container inspection apparatus.
6 . The method according to claim 1 , wherein the at least one statistical evaluation variable is determined based on the set of test evaluation variables while simultaneously inspecting containers based on the set of real-time evaluation variable in the working mode in the container inspection apparatus.
7 . The method according to claim 1 , wherein in the working mode an inspected container is rejected by a rejection device as a function of the evaluation carried out by the real-time evaluation device, and wherein the statistical evaluation variable is characteristic of a statistical rejection variable of the rejection device.
8 . The method according to claim 1 , wherein the statistical evaluation variable is determined locally separately from the container inspection apparatus.
9 . The method according to claim 1 , wherein criteria for storing spatially resolved sensor data on the non-volatile storage device can be preset by a user, in particular a storage period within which the spatially resolved sensor data are to be stored and/or a container type for which spatially resolved sensor data are to be stored and/or a container number for which spatially resolved sensor data are to be stored.
10 . The method according to claim 1 , wherein the set of real-time evaluation parameters currently set in the real-time evaluation device is transmitted to the setting device and the setting device determines a statistical evaluation variable based thereon, so that the currently configured set of real-time evaluation parameters is comparable with a different set of test evaluation parameters.
11 . The method according to claim 1 , wherein the setting device repeatedly determines statistical evaluation variables, wherein further spatially resolved sensor data and/or changes to at least one test evaluation parameter are taken into account.
12 . The method according to claim 1 , wherein a second set of test evaluation parameters to be evaluated is provided to the setting device and a second statistical evaluation variable is determined as a function of this second set of test evaluation parameters on the basis of the retrieved plurality of spatially resolved sensor data.
13 . The method according to claim 1 , wherein the setting device exchanges data with the real-time evaluation device and/or the at least one sensor device via a wireless communication connection, and preferably at least in portions via a public network.
14 . The method according to claim 1 , wherein the real-time evaluation parameters and/or the test evaluation parameters are selected from a group comprising a variable characteristic of at least one ROI, of a sensor setting, of an evaluation accuracy, of a variable characteristic of at least one class of a container state, and similar.
15 . A container inspection apparatus for inspecting containers, having at least one transport device configured for transporting containers to be inspected as a container stream along a predetermined transport path in a working mode, having at least one sensor device configured for detecting, in particular optically, spatially resolved sensor data with respect to a container to be inspected of the container stream in the working mode, and having a real-time evaluation device which is configured for evaluating the spatially resolved sensor data of the individual inspected containers in real time in the working mode with the aid of an adjustable set of real-time evaluation parameters, and with a non-volatile storage device, on which a plurality of the spatially resolved sensor data recorded by the sensor device are stored on a non-volatile storage device during the working mode, wherein, in a setting mode, the plurality of stored spatially resolved sensor data are retrieved by a setting device, wherein a set of test evaluation parameters can be provided to the setting device for evaluation with respect to use as a set of real-time evaluation parameters in the real-time evaluation device, wherein the setting device for evaluating the set of test evaluation parameters is configured for determining a statistical evaluation variable on the basis of the retrieved plurality of spatially resolved sensor data.
16 . The container inspection apparatus according to claim 15 , wherein the setting device is arranged at least partially locally separated with respect to the real-time evaluation device and/or the sensor device.
17 . The container inspection apparatus according to claim 16 , wherein the container inspection apparatus has a rejection device which is configured for rejecting an inspected container as a function of the evaluation carried out by the real-time evaluation device, and wherein the statistical evaluation variable is characteristic of a statistical rejection variable of the rejection device.
18 . The container inspection apparatus according to claim 16 , wherein the transport device is configured to transport the containers from a first treatment device to a second treatment device and preferably the first and/or the second treatment device is selected from a group comprising a cleaning device for cleaning the containers, a filling device for filling the containers, a forming device for forming a plastic preform into a plastic container, in particular a blow molding machine, a labeling device and similar, and a combination thereof.
19 . The container inspection apparatus according to claim 15 , wherein the container inspection apparatus has a rejection device which is configured for rejecting an inspected container as a function of the evaluation carried out by the real-time evaluation device, and wherein the statistical evaluation variable is characteristic of a statistical rejection variable of the rejection device.
20 . The container inspection apparatus according to claim 15 , wherein the transport device is configured to transport the containers from a first treatment device to a second treatment device and preferably the first and/or the second treatment device is selected from a group comprising a cleaning device for cleaning the containers, a filling device for filling the containers, a forming device for forming a plastic preform into a plastic container, in particular a blow molding machine, a labeling device and similar, and a combination thereof.

Description

The present invention relates to a container inspection apparatus for inspecting containers and a method for adjusting, in particular for fine adjustment, a container inspection apparatus. The container inspection apparatus has at least one transport device for transporting containers to be inspected as a container stream along a predetermined transport path, and at least one sensor device, such as a camera, in particular for the optical detection of spatially resolved sensor data, such as image data, in relation to a container to be inspected in the container stream. The containers are preferably plastic containers (in particular PET containers), containers whose main component consists of pulp and/or glass containers and/or cans. The containers may be containers from the beverage and/or food and/or cosmetics industries. For example, they can be cans or bottles, such as glass bottles, pulp bottles and plastic bottles. Such container inspection apparatus has been known from the prior art for some time. An inspection apparatus for inspecting containers and an image evaluation device is known from DE 20 2004 007 783 U1 of the applicant. EP 1 099 948 A2 of the applicant further discloses an apparatus for optical inspection with an image evaluation system, in which an analysis system with an image evaluation program is provided for the analysis of defect images obtained, which substantially corresponds to that of the image evaluation system and with which selected image data are analyzed. In the construction of systems comprising at least one apparatus for the inspection of containers in a container stream, specifications must be observed in accordance with which a guaranteed contamination variable (or more generally a container state variable) is to be discharged. It is also common practice to increase the detection accuracy of the inspection machine or container inspection apparatus until a false rejection rate desired or accepted by the operator is reached. The inspected containers of the container stream are preferably transported to a rejection device, which rejects the containers to be rejected (in accordance with the inspection result). A false rejection rate is understood in particular to be the ratio of containers rejected by the rejection device (in particular undesirably by the operator of the container inspection apparatus), which have no damage or contamination or in particular further undesirable defects or undesirable properties, to a total container intake of the machine (or here the container inspection apparatus). A value of up to 0.1% (per camera unit) is usually tolerated, depending on the inspection unit or container inspection apparatus. However, the total rejection rate cannot be chosen arbitrarily because the inspection machine or container inspection apparatus must produce faster by the amount of rejected containers in order not to negatively influence the line efficiency. The overall rejection rates are normally less than 5%, but can reach 10% for a few seconds. However, such an overall rejection rate suitable for production can generally only be set during production operation with the aid of image evaluation parameters, taking into account the bottle quality to be processed. This means that after the construction of a system and after the start of production, a technician must actually remain on location for a longer period of time in order to adjust the image evaluation parameters in such a manner that the maximum achievable detection accuracy is achieved, taking into account the maximum specified false rejection rate. This is particularly true in view of the fact that a container treatment system almost always has to process several container types and the parameterization process must be carried out individually for each type, taking the rate into account. It is known from the internal prior art that a camera system runs a set of sensitivity parameters in manufacturing (real time). At the same time, ROI (region of interest) changes, parameter (changes) and sensitivity changes can be made without affecting the real-time part. The necessary images can either be “fetched” from the current manufacturing process or from a storage that has previously been loaded with production images. Only when the configurator decides that the parameterization is appropriate, he saves it, i.e., the “offline” parameterization is transferred to manufacturing. How these parameters behave, in particular with regard to false rejection (rate), can only be determined during production, because with a target false rejection rate of <0.1% (per camera), more than 5000 containers (for example approximately 5000-30000 containers in manufacturing) should be inspected in order to be able to recognize a trend. Image processing in inspection machines or container inspection apparatus currently has to be adapted to each production type (of containers produced) by adjusting (evaluation) parameters. Such an adjustment i