Package quality inspection station and package quality inspection method

Abstract

A package quality inspection station (100) is provided. The station (100) comprises a first conveyor belt (102) and a second conveyor belt (108) arranged to contact a bottom (104) of a package (106). The second conveyor belt (108) is arranged at a distance (D) from the first conveyor belt (102) such that a gap (110) is formed. A belt brake (112) comprising first and second side belt arrangements (114, 116) arranged to hold the package (106) is provided to bridge the gap (110) and hold the package (106) such that the top portion (160) and the bottom portion (104) are visible as it is conveyed across the gap (110). The top camera (132) is configured to capture top image data (134) describing the top portion (160), and the bottom camera (136) is configured to capture bottom image data (138) describing the bottom portion (104). The controller (142) is configured to receive the top image data (134) and the bottom image data (138) and to identify quality issues associated with the package (106) based on a combination of the top image data (134) and the bottom image data (138).

Description

[Technical field]

[0001] The present invention relates generally to food packages, and more particularly to a package quality inspection station and method. [Background technology]

[0002] Currently, it is common in the packaging industry to continuously evaluate the quality of the packages produced. In its simplest form, this can be achieved by having personnel evaluate the quality of sample packages. Although this method has proven to be a reliable quality evaluation method in some cases, there is a risk that different personnel will give different evaluations for the same quality issue, which may result in doubting the reliability of the quality evaluation.

[0003] To overcome the risks associated with manual quality assessment, automated quality assessment may also be an option. Instead of having a human to assess the packages, an automated quality station may include a camera that captures image data depicting the package and a controller configured to capture the image data so that deviations associated with a particular package can be detected. For example, if a cap is not properly applied compared to other packages that have been verified to have had the capping process performed according to pre-defined quality standards, this will be detected by the automated quality station and a notification will be sent to an operator. The operator can then manually perform a quality inspection and check the cap application process to verify whether the cap was properly applied.

[0004] Although currently available camera-based quality control stations have improved the consistency of packaging quality control, there is a need for better quality control stations that can detect quality issues earlier and more reliably. Earlier detection of deviations and increased reliability has a positive effect on cost reduction; in other words, detecting improper packaging earlier means less packaging is wasted. Improved quality control means less food is wasted2 and improved quality control stations also have a positive effect on CO2 emissions.

[0005] Patent document TWM542139U discloses an image-based inspection system with multiple cameras for inspecting bottles while they are being transported in a transport unit.

[0006] Patent document US6158193A discloses a system for checking packages, in particular cigarette packs, using a camera to sort out defective packs. Summary of the Invention [Problem to be solved by the invention]

[0007] It is an object of the present invention to at least partially overcome one or more of the above-mentioned limitations of the prior art, and in particular to provide a quality control inspection station that can quickly and reliably detect quality problems in packages, such as carton packages.

[0008] According to a first aspect, a first conveyor belt positioned to contact a bottom portion of the package as it is conveyed along the conveying path; a second conveyor belt positioned to contact a bottom portion of the package as it is conveyed along the conveying path; wherein the second conveyor belt may be spaced apart from the first conveyor belt such that a gap is formed; a belt brake including a first side belt arrangement and a second side belt arrangement arranged to hold the package via a first side section and a second side section of the package as it is conveyed along the conveying path; wherein in a first conveyor section of the conveying path, the packages are conveyed only by the first conveyor belt, in a first transfer section of the conveying path, the packages are conveyed by the first conveyor belt in combination with a belt brake, in a belt brake section of the conveying path, the packages are conveyed only by the belt brake, in a second transfer section of the conveying path, the packages are conveyed by the belt brake in combination with a second conveyor belt, and in a second conveyor section of the conveying path, the packages are conveyed only by the second conveyor belt; an upper camera configured to capture upper image data depicting a top of the package, the upper camera being positioned above the transport path; a bottom camera configured to capture bottom image data depicting a bottom of the package, the bottom camera being disposed in a gap between the first conveyor belt and the second conveyor belt below the belt brake section; a controller configured to receive the top image data and the bottom image data and identify a quality issue associated with the package based on a combination of the top image data and the bottom image data; A package quality inspection station is provided.

[0009] Advantageously, by having a gap and a belt brake to hold the package, image data of the top and bottom can be captured simultaneously or within a short time period, for example within 3 seconds, more preferably within 1 second. The effect is that the top and bottom portions can be evaluated simultaneously. This is advantageous because for carton packages formed from blanks or webs, misalignment of the top can lead to misalignment of the bottom and vice versa.

[0010] In other words, the belt brake is provided to hold the package such that the top and bottom of the package are accessible to the top and bottom cameras, respectively. As mentioned above, this is possible because the belt brake comprises first and second side belt arrangements that can hold the package by interacting with the sides of the package. Thus, in this specification, the belt brake is not necessarily used to slow down the package being fed through the belt conveyor, but rather to hold the package such that the top and bottom of the package are within the field of view of the top and bottom cameras, respectively.

[0011] As mentioned above, packages formed from webs and packages formed from blanks may benefit from the quality inspection system described above. Some quality issues with such packages are likely to be found on both the top and bottom of the package. For example, if the folding of the package is done in such a way that the flaps at the top of the package do not align, this may result in the flaps at the bottom of the package not aligning. This is relevant as the packaging material is folded from a rectangular or other pre-set shape of packaging material. In other words, the features of the top are linked to the features of the bottom and vice versa, so it has proven beneficial to use the image data of the top and bottom in combination to identify quality issues with the package, i.e., by evaluating the two image data with the knowledge that they relate to the same package. In doing so, models, e.g. statistical models and AI / ML (artificial intelligence / machine learning) based models, have the positive effect of being able to perform more reliable quality inspections.

[0012] Since some packages, e.g. carton packages, may be deformed such that the top and / or bottom features are affected, it is beneficial to provide that the top image data and the bottom image data are captured simultaneously, or if simultaneous capture is not possible, at least within a short time span, e.g. less than 3 seconds. By reducing the time difference between capturing the top image data and capturing the bottom image data, there is less risk that the top and / or bottom will be affected in any way between capturing the two different image data sets. Thus, by capturing the two image data sets within a short time or simultaneously, the correlation between the top and bottom features can be determined more reliably.

[0013] The top camera may be positioned vertically above the bottom camera.

[0014] A photocell may be provided upstream of the transport path and may trigger the capture of top image data using the top camera and / or bottom image data using the bottom camera.

[0015] The controller may include a text recognition module configured to convert text included in the upper image data into a text data set.

[0016] The controller may further include an identification code recognition module configured to convert an identification code depicted in the upper image data into an identification data set.

[0017] Being able to link an identification data set to a package has the advantage that the controller's identification of quality issues can benefit from other information linked to the package, e.g. machine data, i.e. the settings used in the filling machine (also called packaging machine) used when producing the package.

[0018] The controller may further include an upper portion formation evaluation module configured to evaluate an upper portion feature, such as a top flap, provided on a top portion of the package depicted in the upper portion image data.

[0019] The controller may further include a bottom formation assessment module configured to assess a bottom feature, such as a bottom flap, provided at the bottom of the package depicted in the bottom image data.

[0020] The top formation evaluation module may be configured to provide top feature evaluation data to the bottom formation evaluation module, which may be arranged to evaluate the bottom features based on the bottom image data in combination with the top feature evaluation data.

[0021] The controller may further include a data communication module configured to transfer the package identification data set, the top image data, and the bottom image data to a database on a server; Wherein, the server may be further configured to receive operator feature evaluation data and an identification data set from the operator panel, and link the operator feature evaluation data to the top image data and the bottom image data via the identification data set, and the operator feature evaluation data may specify whether the package meets the feature quality requirement with respect to the top feature; As a result, a combination of the top image data, bottom image data and operator feature assessment data may be used as feature learning data for an artificial intelligence engine used by the top formation assessment module and / or the bottom formation assessment module.

[0022] By combining the top and bottom image data with the operator feature evaluation, the automatically captured data can be efficiently combined with the manual quality assessment data. The advantage of this is that data from different sources can be used, for example, by an AI engine, resulting in a more reliable quality inspection.

[0023] The upper camera may be a 3D camera configuration that generates 3D cap image data, and the controller may include a cap application evaluation module configured to evaluate a cap applied to an upper portion of the package depicted in the upper image data.

[0024] The data communication module may be configured to transfer the 3D cap image data to a database on the server; Wherein, the server may be further configured to receive operator cap evaluation data from the operator panel; and link the operator cap evaluation data to the 3D cap image data, wherein the operator cap evaluation data identifies whether the application of the cap on the package meets the cap application quality requirements; The combination of the 3D cap image data and the operator cap evaluation data may be used as cap application training data for the AI ​​engine used by the cap application evaluation module.

[0025] According to a second aspect, there is provided a method for inspecting quality of a package, the method comprising: conveying the package along a conveying path onto the first conveyor belt by placing the package on the first conveyor belt such that a bottom of the package rests on the first conveyor belt; transferring the package from the first conveyor belt to a belt brake comprising the first and second side belt arrangements by having the bottom rest on the first conveyor belt and first and second side portions of the package interacting with the first and second side belt arrangements, respectively, to transport the package along a transport path; interacting the package with the first and second belt arrangements to transport the package across a gap formed between the first conveyor belt and a second conveyor belt disposed downstream of the first conveyor belt in the transport path; capturing top image data depicting a top of the package using the top camera during the step of transporting the package across the gap; capturing bottom image data describing a bottom of the package using a bottom camera during the step of transporting the package across the gap; transferring the package from the belt brake to the second conveyor belt by conveying the package along a conveying path with the bottom resting on the second conveyor belt and first and second side sections of the package interacting with the first and second side belt arrangements, respectively; conveying the package along the conveying path onto the second conveyor belt by placing the package on the second conveyor belt such that a bottom of the package rests on the second conveyor belt; providing the top image data and the bottom image data to a controller; using the controller to combine the top image data and the bottom image data to identify quality issues associated with the package; Prepare.

[0026] The features and advantages mentioned in relation to the first aspect also apply to the second aspect.

[0027] Further objects, features, aspects and advantages of the present invention will become apparent from the following detailed description and drawings. [Means for solving the problem]

[0028] Embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings, in which: FIG. [Brief description of the drawings]

[0029] [Figure 1A] 1 shows a schematic side view of a package quality inspection station. [Figure 1B] 1 shows a schematic top view of a package quality inspection station. [Diagram 2] A top view of the package is shown. [Diagram 3] A bottom view of the package is shown. [Figure 4]1 shows a flow chart illustrating a method for inspecting the quality of a package. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0030] 1A and 1B, there is shown an example package quality inspection station 100. Figure 1A shows a side view of package quality inspection station 100, and Figure 1B shows a top view thereof.

[0031] The package quality inspection station 100 comprises a first conveyor belt 102 arranged to interact with the bottom 104 of the package 106. As shown, the first conveyor belt 104 may be belt-driven, but other techniques for transporting the package 106 along the transport path P may be used. Downstream of the first conveyor belt 104, a second conveyor belt 108 is provided. The first and second conveyor belts 102, 108 are arranged at a distance D from each other such that a gap 110 is formed. To transport the package 106 across the gap 110, a belt brake 112 is used. It comprises a first side belt arrangement 114 and a second side belt arrangement capable of clamping the package 106. More specifically, the package 106 may be held by the belt brake 112 by a first side section 118 of the package 106, such as a first side panel, interacting with the first side belt arrangement 114 while a second side section 120 of the package 106, such as a second side panel, interacts with the second side belt arrangement 116. Stated another way, the package 106 is clamped between the first side belt arrangement 114 and the second side belt arrangement 116 such that the package 106 is lifted across the gap 110.

[0032] To ensure a smooth transition from the first conveyor belt 102 to the belt brake 110 and from the belt brake 110 to the second conveyor belt 108, the belt brake 110 extends over the first conveyor belt 102 as well as the second conveyor belt 108. In other words, as shown, in the first conveyor section 122, the package 106 can be placed on the first conveyor belt 102 without interacting with the belt brake 110. In the first transfer section 124 located downstream of the first conveyor section 122, the package 106 comes into contact with the belt brake 110 such that the bottom 104 of the package 106 comes into contact with the first conveyor belt 102 and the first and second side sections 118, 120 of the package 106 can come into contact with the first and second side sections 118, 120, respectively. Next, in a belt brake section 126 located downstream of the first transfer section 124, the package 106 may be held only by the belt brake 112, more specifically the first and second side belt arrangements 114, 116, thereby providing a bottom 104 of the package 106 visible from below.

[0033] To ensure that the packages 106 are transferred smoothly and reliably from the belt brake 112 to the second conveyor belt 108, a second transfer section 128 is provided. In this section, the packages 106 are held by the belt brake 112 in combination with being placed on the second conveyor belt 108. Downstream of the second transfer section 128, a second conveyor section 130 is provided. In this second conveyor section 130, the packages 106 are transported only by the second conveyor belt 108.

[0034] Once mounted in the belt brake section 126, as shown in more detail in FIG. 2, the package 106, more particularly the top surface portion 160 of the package 106, may be positioned within the field of view of the upper camera 132 so that upper image data 134 may be generated. The upper image data 134 may be two-dimensional image data, but may also be three-dimensional image data. In the latter case, an additional camera 133 may be used. However, since the speed of the belt brake may be known, it is also possible to use the upper camera 132 to capture image data at different times and form the three-dimensional image data by combining the image data from the different times into the three-dimensional image data by utilizing the known speed of the belt brake 112.

[0035] The top camera 132 and the additional top camera 133 may be configured in a similar manner, but they may also differ, by way of example, in that they are configured to capture light having different spectral characteristics, i.e., have different filters and / or image sensors configured to capture light of different spectral characteristics. Furthermore, as shown, the top camera 132 and the additional top camera 133 may be tilted differently, thereby providing, for example, that reflections on the package 106 are depicted differently in the image data generated by the top camera 132 compared to the additional top camera 133. As shown, the top camera 132 may be tilted at an angle α, and the additional top camera 133 may be tilted at an angle β, where the angle β is greater than the angle α. Furthermore, the top illumination device 135 is used to provide conditions for the top camera 132 and, optionally, the additional top camera 133 to generate top image data 134 that can be used as a basis for making a reliable quality assessment.

[0036] The bottom camera 136 may be located below the belt brake section 126. Since the package 106 is held only by the belt brake 112 and the bottom section 104 is not resting on either the first conveyor belt 102 or the second conveyor belt 108, it becomes possible to capture bottom image data 138 depicting the bottom section 104 of the package 106. The advantage of locating the bottom camera 136 as well as the top camera 132 and optionally the additional top camera 133 in the same section is that the package 106 has the same condition when assessing the top section 160 and the bottom section 104. Having the possibility to assess these two at approximately the same time provides the advantage that dents in the side panels and other quality issues that may be present in the top and bottom sections are easier to detect compared to when the top section 160 and the bottom section 104 are assessed at different times, even if they do not necessarily affect these sections directly. The risk of evaluating the top 160 and bottom 104 at different times is that the package 106 has been affected between these two different times and therefore the condition of the package is not the same. Furthermore, a bottom illumination device 139 is used to provide conditions that increase the likelihood of obtaining bottom image data 134 that can be used as a basis for quality assessment.

[0037] As shown, a photocell 140 may further be provided upstream of the top camera 132 and the bottom camera 136, which allows triggering the capture of the top image data 134 and the bottom image data 138 by sending a trigger signal to the top camera 132, the optional additional top camera 133, and the bottom camera 136. Since the distance between the photocell 140 and the top camera 132 and the bottom camera 136, as well as the speeds of the first conveyor belt 102, the belt brake 112, and the second conveyor belt 108, are known, the top camera 135 and the bottom camera 136 may be adjusted to capture the top image data 134 and the bottom image data 138 at the time when the package 106 is located at the position of the transport path P where the top camera 135 and the bottom camera 136 are configured. Although the photocell 140 is taken as an example, this should be considered as one way of triggering the image capture, and other technologies that fulfill the task of detecting the package 106 may be used as an alternative to the photocell 140.

[0038] The data generated by the top camera 132 and the bottom camera 136 may be provided to a controller 142, which may be any data processing device capable of handling the data and may be embodied as a single unit or multiple units. The controller 142 may include a text recognition module 144 configured to receive the top image data 134 and convert text 146 on the package 106 depicted in the top image data 134 into a text data set.

[0039] The controller 142 may further include an identification code recognition module 150 configured to receive, via the upper image data 134, an identification code 152, such as a QR code, printed on the upper portion 160 of the package 106. Based on the identification code 152, an identification data set 154 is determined. For example, the identification code 152 may be embodied as a QR code or a code consisting of a combination of letters and numbers, as shown in FIG. 2. The identification code recognition module 150 interprets this code depicted in the upper image data 134 and converts it into an identification data set 154, which may be in the form of a number. Since the identification data set 154 is in the form of a number, the identification data set 154 can be easily compared with other data and can be efficiently stored.

[0040] Additionally, the controller 142 may include a top formation assessment module 156 configured to receive top image data 134 depicting top features 158. Based on the top image data 134, locations, dents, etc. on the top features 158, illustrated here as top flaps, may be identified and determined to be quality issues if the locations, dents, etc. are deemed to be deviations from quality standards. As described further below, the quality standards may be based on data collected from previously assessed packages.

[0041] Since bottom image data 138 is also available, a bottom formation assessment module 162 may also be provided. The bottom image data 138 may depict bottom features 164 as shown in FIGURE 3, and similar to the top formation assessment module 156, the bottom formation assessment module 162 is configured to look at the bottom features 164 and detect deviations compared to quality standards. Similar to the top formation assessment module 156, the bottom formation assessment module 162 may compare the bottom features 164 of a sample package, i.e., the package being assessed, to previously assessed packages that were determined to meet the quality standards.

[0042] The evaluation of the top feature 158 and the bottom feature 164 may be interrelated, especially when the package 106 is a carton package produced by a roll-fed wrapper as shown in Figures 2 and 3. Having a package 106 produced by a roll-fed wrapper entails that misalignment at one end of the package 106, e.g., the top portion 160, may also cause misalignment at the other end, e.g., the bottom portion 104. The top feature evaluation data 166 may be transferred from the top formation evaluation module 156 to the bottom formation evaluation module 162 to provide that the evaluation made looking at the top features is used when evaluating the bottom features. Although not shown, the reverse may also be done, i.e., the bottom feature evaluation data may be transferred from the bottom formation evaluation module 162 to the top formation evaluation module 156. Sharing data in this manner further improves the reliability of the quality evaluation and takes into account interrelationships between the top 160 and the bottom 104 that are not considered in today's quality inspection stations.

[0043] In this manner, the top and bottom image data 134, 138 are evaluated in combination when identifying quality issues with the package 106 based on the relationship between the top and bottom of the package 106. In other words, it is taken into consideration that the two sets of image data relate to the same package 106, and thus a quality issue with the top portion 160 may be correlated with a quality issue with the bottom portion 104.

[0044] The controller 142 may include a data communication module 170. The advantage of having this module is that data can be exchanged with other devices, making it possible to utilize information from a number of different package quality inspection stations 100. To further improve the quality assessment, information collected via the operator panel 174 can also be taken into account. For example, the top image data 134, the bottom image data 138, and the identification data set 154 can be transferred from the controller 142 to a server 176 including a database 178 using the data communication module 170. In parallel, the operator top feature assessment data 172, i.e. data relating to the assessment of the top feature 158 made by the operator, and the identification data set 154 may be transferred from the operator panel 174 to the server 176 and the database 178. Since the identification data set 154 is provided both via the controller 142 and the operator panel 174, the operator feature assessment data 172 can be linked to the top image data 134 and the bottom image data 138 provided via the controller 142. By having the possibility to link data provided via the controller 142 and the operator panel 174, the database 178 can include not only automatically captured data but also manually captured data. The advantage of having data captured from both the top camera 132 and the bottom camera 136 and the operator panel 174 is that different aspects of the package 106 are covered and an improved basis for further analysis can be formed. By using an AI engine 180 such as a neural network, the data provided to the server 176 in combination with data that may be provided to the server 176 from the operator panel 174 as a quality approval or disapproval, i.e. that the package 106 meets or does not meet a quality standard, can be used to continuously improve the model used to evaluate the top feature 158 and / or the bottom feature 164 as well as other parts of the package 106.

[0045] If the package 106 includes a cap 185, the top camera may be a 3D camera arrangement that generates 3D cap image data 182. The 3D cap image data 182 may be transferred to a cap application evaluation module 184 on the controller 142, where the application of the cap 185 is evaluated, for example, whether the cap is applied to an assigned area on the top of the package. The 3D cap image data 182 may also be transferred from the controller 142 to the server 176. The application of the cap may then be evaluated by an operator, via the operator panel 174, and operator cap evaluation data 186 may be transferred to the server 176, where the AI ​​engine 180 may be trained by the 3D cap image data 182 and the operator cap evaluation data 186.

[0046] FIG. 2 exemplarily illustrates a package 106 embodied as a brick-shaped carton package. As illustrated, the package 106 may include a first side panel forming a first side section 118 and a second side panel forming a second side section 120. In this particular example, a top feature 158, shown herein as a top flap, is folded over the first and second side sections 118, 120. By tilting the top camera 132 as shown in FIG. 1A, the top image data 134 may cover not only the top panel but also a portion of the side panel of the package 106 such that the top feature 158 may be evaluated even when folded over the side panel as shown in FIG. 2, which follows the commonly known folding method of carton packages, such as the Tetra Brik® package sold by Tetra Pak®.

[0047] The text 146 is depicted in the example illustrated in Figure 2 as a code relating to a best before date and a filler identification code. The identification code 152 is illustrated as a code consisting of a combination of letters and numbers. As mentioned above, the code may also be provided in the form of a QR code or other two-dimensional code or other non-textual format.

[0048] The cap 185 is shown in this example as a combination lid provided on a base. The base may be glued to the package 106. The position of the base may form part of the features evaluated in the quality assessment to ensure there is no leakage or misalignment between the base of the cap and the package. A straight edge of the base may be an advantage as it may facilitate alignment-related comparisons, for example to assess whether the cap 185 is properly aligned with the edge of the top panel of the package 106.

[0049] FIG. 3 illustrates the bottom 104 of the package 106. Similar to the example illustrated in FIG. 2, the package 106 illustrated in FIG. 3 is a carton package produced by a roll-fed filling machine. Unlike the folding of the top 160 illustrated in FIG. 2, the bottom feature 164, here illustrated by the bottom flap, is folded inward, toward the bottom panel, so that the bottom flap is not visible from the side view of the package 106. Just as the folding principle of the top is generally known, the same is true for the folding of the bottom 104 illustrated in FIG. 3. However, the impact of the bottom feature being folded inward is that quality assessments that do not consider the visual appearance of the bottom 104, more specifically the part of the package 106 that is arranged to contact the underlying surface, cannot evaluate the bottom feature 164 in the same way and are potentially less reliable, compared to the package quality inspection station 100 illustrated in FIG. 1A and FIG. 1B.

[0050] FIG. 4 is a flow chart illustrating a method 200 for quality inspection of a package 106 using, for example, the package quality inspection station 100 shown in FIGS. 1A and 1B.

[0051] In a first step 202, the package is conveyed on the first conveyor belt 102. Then, in a second step 204, the package 106 is transferred from the first conveyor belt 102 to the belt brake 112 by conveying the package 106 on the first conveyor belt 102 in combination with the belt brake 112, as described above. After the transfer, in a third step 206, the package 106 may be conveyed across the gap 110 formed between the first conveyor belt 102 and the second conveyor belt 108. While being conveyed across the gap 110, in a fourth step 208, the top camera 132 is used to capture top image data 134. Also while being conveyed across the gap 110, in a fifth step 210, the bottom camera 136 is used to capture bottom image data 138. Thereafter, in a sixth step 212, the package 106 is handed over from the belt brake 112 to the second conveyor belt 108. After the handover, in a seventh step 214, the package 106 may be transported by the second conveyor belt 108. In an eighth step 216, the top image data 134 and the bottom image data 138 may be provided to a controller 142, which may include a processor and memory, so that the data may be processed. In a ninth step 218, quality issues associated with the package 106 are identified based on a combination of the top image data 134 and the bottom image data 138.

[0052] Even if a particular order is described, other orders may be applied.

[0053] From the foregoing description, while various embodiments of the present invention have been described and illustrated, the invention is not limited thereto and may be embodied in other ways within the scope of the subject matter defined in the following claims.

Claims

1. a first conveyor belt (102) arranged to abut against a bottom (104) of the package (106) as it is conveyed along the conveying path (P); a second conveyor belt (108) arranged to abut the bottom (104) of the package (106) as it is conveyed along the conveying path (P); the second conveyor belt (108) is spaced a distance (D) from the first conveyor belt (102) such that a gap (110) is formed; a belt brake (112) comprising a first side belt arrangement (114) and a second side belt arrangement (116) arranged to hold the package (106) via a first side section (118) and a second side section (120) of the package (106) as it is conveyed along the conveying path (P); In a first conveyor section (122) of the conveying path (P), the package (106) is conveyed only by the first conveyor belt (102); in a first transfer section (124) of the conveying path (P), the package (106) is conveyed by a combination of the belt brake (112) and the first conveyor belt (102); in a belt brake section (126) of the conveying path (P), the package (106) is conveyed only by the belt brake (112); in a second receiving section (128) of the conveying path (P), the package (106) is conveyed by a combination of the belt brake (112) and the second conveyor belt (108); in a second conveyor section (130) of the conveying path (P), the package (106) is conveyed only by the second conveyor belt (108); an upper camera (132) disposed above the transport path (P) and configured to capture upper image data (134) depicting an upper portion (160) of the package (106); a bottom camera (136) disposed below the belt brake section (126) in a gap (110) between the first conveyor belt (102) and the second conveyor belt (108), the bottom camera (136) configured to capture bottom image data (138) depicting a bottom (104) of the package (106); a controller (142) configured to receive the top image data (134) and the bottom image data (138) and to identify quality issues associated with the package (106) based on a combination of the top image data (134) and the bottom image data (138); A package quality inspection station (100).

2. The top camera (132) is disposed vertically above the bottom camera (136).

2. The package quality inspection station (100) of claim 1.

3. a photocell (140) is provided upstream of the top camera (132) or the bottom camera (136) on the transport path (P) to trigger the capture of the top image data (134) using the top camera (132) or the capture of the bottom image data (138) using the bottom camera (136); 2. The package quality inspection station (100) of claim 1.

4. the controller (142) includes a text recognition module (144) configured to convert text (146) included in the upper image data (134) into a text data set; 2. The package quality inspection station (100) of claim 1.

5. the controller (142) includes an identification code recognition module (150) configured to convert an identification code (152) depicted in the upper image data (134) into an identification data set (154); 2. The package quality inspection station (100) of claim 1.

6. the controller (142) includes an upper portion formation evaluation module (156) configured to evaluate an upper portion feature (158), such as an upper portion flap, on an upper portion (160) of the package (106) depicted in the upper portion image data (134); 2. The package quality inspection station (100) of claim 1.

7. the controller (142) includes a bottom formation evaluation module (162) configured to evaluate a bottom feature (164), such as a bottom flap, on a bottom (104) of the package (106) depicted in the bottom image data (138); 2. The package quality inspection station (100) of claim 1.

8. the top formation assessment module (156) is configured to provide the top feature assessment data (166) to the bottom formation assessment module (162), the bottom formation assessment module (162) being configured to assess the bottom features (164) based on the bottom image data (138) in combination with the top feature assessment data (166); 8. A package quality inspection station (100) according to claim 7.

9. the controller (142) further comprises a data communication module (170) configured to transfer the package (106) identification data set (154), the top image data (134), and the bottom image data (138) to a database (178) on a server (176); The server (176) is further configured to receive operator feature evaluation data (172) and the identification data set (154) from an operator panel (174), and link the operator feature evaluation data (172) to the top image data (134) and the bottom image data (138) via the identification data set (154), the operator feature evaluation data (172) identifying whether the package (106) meets feature quality requirements with respect to the top feature (158); a combination of the top image data (134), the bottom image data (136) and the operator feature assessment data (172) is used as feature learning data for an artificial intelligence (AI) engine (180) used by the top formation assessment module (156) and / or the bottom formation assessment module (162); 9. A package quality inspection station (100) according to claim 8.

10. the upper camera (132) comprises a 3D camera arrangement that generates 3D cap image data (182), and the controller (142) comprises a cap application evaluation module (184) configured to evaluate a cap (185) applied to a top portion (160) of the package (106) depicted in the upper image data (134).

2. The package quality inspection station (100) of claim 1.

11. the data communication module (170) is configured to transfer the 3D cap image data (182) to a database (178) on the server (176); the server (176) is further configured to receive operator cap evaluation data (186) from the operator panel (174) and link the operator cap evaluation data (186) to the 3D cap image data (182); the operator cap evaluation data (186) identifying whether application of a cap (185) onto the package (106) meets cap application quality requirements; a combination of the 3D cap image data (182) and the operator cap evaluation data (186) is used as cap application training data for an AI engine (180) used in the cap application evaluation module (184); The package quality inspection station (100) of claim 10.

12. A method (200) for quality inspection of a package (106), the method comprising: conveying (202) the package (106) onto the first conveyor belt (102) along a conveying path (P) by placing the package (106) on the first conveyor belt (102) such that a bottom (104) of the package (106) rests on the first conveyor belt (102); said bottom (104) resting on said first conveyor belt (102), said first and second side sections (118) and (120) of said package (106) interacting with said first and second side belt arrangements (114) and (116), respectively, to transport said package (106) along said conveying path (P), and transferring (204) said package (106) from said first conveyor belt (102) to a belt brake (112) comprising said first and second side belt arrangements (114) and (116); conveying (206) the package (106) across a gap (110) formed between the first conveyor belt (102) and the second conveyor belt (108) disposed downstream of the first conveyor belt (102) in the conveying path (P) by interaction of the package (106) with the first side belt arrangement (114) and the second side belt arrangement (116); capturing (208) top image data (134) depicting a top portion (160) of the package (106) using a top camera (132) during a step (206) of conveying the package (106) across the gap (110); During the step of conveying (206) the package (106) across the gap (110), capturing (210) bottom image data (138) depicting a bottom (104) of the package (106) using a bottom camera (136); the bottom (104) rests on the second conveyor belt (108), the first side portion (118) and the second side portion (120) of the package (106) interact with the first side belt arrangement (114) and the second side belt arrangement (116), respectively, to transport the package (106) along the conveying path (P) and to hand over (212) the package (106) from the belt brake (112) to the second conveyor belt (108); conveying (214) the package (106) along the conveying path (P) onto the second conveyor belt (108) by placing the package (106) on the second conveyor belt (108) such that the bottom (104) of the package (106) rests on the second conveyor belt (108); providing (216) the top image data (134) and the bottom image data (138) to a controller (142); using the controller (142) to combine the top image data (134) and the bottom image data (138) to identify (218) quality issues associated with the package (106); How to prepare for this.

Images