Full automated container inspection

Abstract

An automated inspection system (2) for containers (6) of different size and weight, comprising of at least one digital camera system (5) which can be mounted in different positions, a rotatable platform (3) for positioning the containers (6) so they can be scanned by the at least one digital camera system (5), wherein either the at least one digital camera system (5) or the rotatable platform (3) or both are adjustable in height, a container transport system (7) which transports the containers (6) to the platform (3), a computer (4) which is connected to the at least one digital camera system (5) and which has access to a database and a software (8) performed by the computer (4) which analyzes digital images (10) taken by the at least one digital camera system (5) and determines the quality of the scanned container (6).

Description

[0001] FULL AUTOMATED CONTAINER INSPECTION

[0002] The hereby described invention discloses a method and a system for an automated inspection of containers of different size and weight.

[0003] Technical Field

[0004] The invention deals with the technological area of quality management in the semiconductor industry.

[0005] Background and description of the prior art

[0006] In the semiconductor industry there are more than 50,000 known product transport containers of various sizes and shapes in circulation. When these containers are returned from the customer, they will be emptied, cleaned, conditioned, refilled with product in a controlled manner using the existing systems in our production site and send to the customer again. Various points have to be checked before delivery:

[0007] 1 . Which container and what contents? - usually it is done via Data Matrix codes

[0008] 2. Are the correct product and safety labels attached in the right place?

[0009] 3. Are all connections sealed?

[0010] 4. Are the closures secured with cotter pins?

[0011] 5. Is the pipework, valves and connections in order and not bent?

[0012] 6. Is the container not damaged?

[0013] The inspection of product transport containers in the Electronics Semiconductor industry is therefore a critical process to ensure the integrity and safety of the products during transit. Prior to the advent of advanced automation and Al technologies, this process was largely manual, involving significant human input and intervention. Traditionally, container inspection involved visual checks and manual measurements by operators to verify various aspects such as the correct attachment of labels, the integrity of seals and closures, and the overall condition of the container. This method was time-consuming and prone to human error, often leading to inconsistencies in quality control and increased rates of customer complaints.

[0014] Documentation of the inspection results was also a manual process. Operators would record their findings on paper, leading to a paper-based filing system. Such systems were not only inefficient but also susceptible to loss, damage, and misfiling, making retrieval and analysis of historical data difficult and error-prone.

[0015] Weighing of containers to determine the remaining quantity of product was traditionally done using standalone industrial scales. These scales required manual operation and the recording of weight data, which then needed to be manually cross-checked against expected weights to verify contents.

[0016] Early camera systems were used in some instances to assist with the manual inspection process. However, these systems were rudimentary and required operators to interpret the camera feed, rather than providing automated analysis. The use of such systems was limited in scope and did not significantly reduce the workload on operators or the potential for error.

[0017] Data management in the context of container lifecycle was minimal and primarily paper-based. Electronic systems, where they existed, were rudimentary and did not provide the level of detail or accessibility required for robust analysis. Integration between weighing systems, camera feeds, and data management systems was also lacking, preventing a seamless flow of information. The limitations of these prior art methods were manifold. The manual nature of inspection and documentation processes was not only labor- intensive but also led to a higher rate of errors. The absence of advanced data management meant that tracking the lifecycle of containers was inefficient, which could lead to difficulties in tracing the history of a container's use and condition over time.

[0018] The technological landscape before the introduction of automated and Al- driven systems left much to be desired in terms of efficiency, reliability, and traceability. The need for improved inspection methods that could handle the volume and complexity of containers used in the semiconductor industry was becoming increasingly apparent, driving the development of more sophisticated solutions.

[0019] Summary of the invention

[0020] The task of this patent application is therefore to provide an improved automated container inspection system which works more reliable and efficient than the known manual and automatic approaches.

[0021] This task has been solved by an automated inspection system for containers of different size and weight, comprising of at least one digital camera system which can be mounted in different positions, a rotatable platform for positioning the containers so they can be scanned by the at least one digital camera system, wherein either the at least one digital camera system or the rotatable platform or both are adjustable in height, a container transport system which transports the containers to the platform, a computer which is connected to the at least one digital camera system and which has access to a database and a software performed by the computer which analyzes digital images taken by the at least one digital camera system and determines the quality of the scanned container. The automated inspection system is designed to handle containers of varying sizes and weights. The key feature is the inclusion of at least one digital camera system, which can be positioned flexibly. The rotatable platform allows the container to be oriented for optimal scanning by the camera(s). The height-adjustability of the camera system or platform (or both) enhances the system’s versatility to accommodate different container dimensions. The container transport system moves the containers into position on the platform. The computer, connected to the camera system, has database access and runs software to analyze images and assess container quality. This configuration streamlines the inspection process, enabling precise image capture for quality determination, thus reducing the risk of human error and increasing throughput. This developed system is furthermore based on various camera applications, linked to a scale that enables fully automated inspection and documentation. It enables a reduction of errors compared to manual inspection of containers, as well as errors in manual documentation and filing on paper, which will reduce customer complaints and costs tremendously. The system furthermore makes it easier to find out the causes of faulty containers.

[0022] Advantageous and therefore preferred further developments of this invention emerge from the associated subclaims and from the description and the associated drawings.

[0023] One of those preferred further developments of the disclosed system comprise that either the digital camera system is used as a reader for a machine-readable code, in particular a data matrix code, which is attached to the container, or the system comprises of at least one additional Code reader device which can be mounted in different positions and is adjustable in height. The system can utilize the digital camera both for imaging and as a reader for machine-readable codes, like a Data Matrix code, on the container. Alternatively, it can include an additional code reader device. Both camera and code reader can be flexibly mounted and height-adjusted. This dual functionality of the camera system or the additional code reader enables efficient identification and tracking of containers, facilitating better inventory management.

[0024] Another one of those preferred further developments of the disclosed system comprise that the computer is a Programmable Logic Controller (PLC) with a touch panel as User Interface. The system uses a Programmable Logic Controller (PLC) as the computer, featuring a touch panel user interface. This allows for robust control of the inspection system and an interactive platform for operators to manage the process. Furthermore the PLC enhances the reliability of the system and simplifies the operator interaction, leading to improved usability and reduced training requirements.

[0025] Another one of those preferred further developments of the disclosed system comprise that a network connects the computer and the at least one digital camera. The PLC can be networked with the digital camera(s), ensuring seamless communication between the control unit and the imaging system. Networking enables real-time data transfer and system coordination, minimizing delays and enhancing the efficiency of the inspection process.

[0026] Another one of those preferred further developments of the disclosed system comprise that the database is either stored on an internal memory or on an external memory which is connected to the computer via the network. The database storing inspection data can be located on internal memory or an external memory accessible via the network, providing flexibility in data storage solutions. This feature allows for scalable data management, accommodating growing data storage needs without compromising system performance.

[0027] Another one of those preferred further developments of the disclosed system comprise that the container are varying in size from from 1 kg up to 400kg. The system is capable of handling a wide range of container weights, from as light as 1 kg to as heavy as 400kg. Such capacity ensures the system’s applicability across a diverse set of container types, making it a versatile solution for various product lines.

[0028] Another one of those preferred further developments of the disclosed system comprise that the at least one digital camera system is attached to a robot which enables the at least one digital camera system to be movable in all axes and to scan the container from all sides. The digital camera system can be mounted on a robot, granting the camera full mobility in all axes, enabling comprehensive scanning of the container from any angle. Robotic mounting allows for thorough inspection of containers, ensuring high-quality control by capturing images from multiple perspectives.

[0029] Another solution for the given task is a method for inspecting containers of different size and weight using a system as previously described, comprising the following steps of Transporting a container to the rotatable platform via the transport system and adjust the height of either the at least one digital camera system or the platform or both depending o size and form of the container; Scanning the container from all sides by taking digital images via the at least one digital camera; Sending the digital image data to the computer; Determining the quality of the scanned container by the software on the computer analyzing the digital image data; and If the determined quality fails a given quality standard labeling it as defect by the software. The quality standard can comprise of many different parameters, e.g. if any visible damage could be detected via the image analyzes, in particular if such damages exceed a specific size or structure. These parameters and values are provided to the software and the computer beforehand. The inspection method involves automated transportation and adjustment of the container and camera system, digital scanning, data transmission, image analysis, and quality determination, with defective containers being marked accordingly. This methodical approach ensures a systematic and thorough inspection process, leading to accurate quality assessment and categorization of containers.

[0030] Another preferred further developments of the disclosed method comprise that the software uses an artifcial neural network to analyze the quality of the containers and to determine a good or bad container. Utilizing a neural network for analysis allows for sophisticated, learning-based assessment that improves over time, reducing false positives and negatives in quality control.

[0031] Another one of those preferred further developments of the disclosed method comprise that the computer creates a documentation of the container inspection comprising all relevant inspection data and stores the documentation in a data archive in the database. This documentation process ensures traceability and accountability, providing a digital history for each container that can be invaluable for quality assurance and customer service.

[0032] Detailed description of the invention

[0033] The method and system according to the invention and functionally advantageous developments of those are described in more detail below with reference to the associated drawings using at least one preferred exemplary embodiment. In the drawings, elements that correspond to one another are provided with the same reference numerals.

[0034] The drawings show:

[0035] Figure 1 : A schematic for an example of the invented container inspection system

[0036] Figure 2: A schematic workflow which applies the invented system Figure 2 shows an overview about the necessary method steps performed by an application which represents a preferred embodiment. The steps itself are performed divergent in every exemplary embodiment dependent on the different conditions. The invented system which performs the method including its components is shown in Figure 1. The shown structure is only exemplary for this preferred embodiment. It can change for a other embodiments. Especially the kind of involved computers can differ greatly, depending on how much of the steps is performed by human users with the help of computers and application software or done automatically by specific computers using for instance Al based software.

[0037] In the preferred exemplary embodiment as shown in Figure 1 a Siemens PLC 4 for overall control of the system 2 is used and two Basler color field cameras 5 for the image analysis. Both systems 4, 5 are communicating via a network 11 to scan, control and archive process data. For the moving platform 3 and cameras 5 use linear actuators from Parker are used. For the ID scanning of the container 6 is a Keyence 2D code scanner installed. The Container Inspection System (CIT) 2 is operated via a HMI Touchpanel 9 from Siemens. For safety reasons, two SIL door sensors are installed that switch off the system 2 when its doors are opened during normal operation. In this preferred exemplary embodiment an image processing application from Matrox (Design Assistant) using neural networks is used by the camera system for image analysis and image recognition of the good and bad containers 6. The results will then be communicated via the network to the control software 8 on the PLC 4. In other exemplary embodiment the image processing application could be also hosted and used by the control software 8 which is performed by the computer or PLC 4. As a documentation a protocol for the quality check of each container 6 is provided in the Unified Siemens control software 8. To collect all data in a central database an OPC communication to an container management system is provided. The operation as hown in Figure 2 is as follows:

[0038] The container 6 will be placed manually or for higher throughput via a robot system on the rotating platform 3 of the CIT 2. For different container types special adapters are available to locate the container 6 in the right position.

[0039] Then a user 1 has to close the doors of the CIT 2 and push the start button. All the checking steps will be done automatically. A green light indicates when the container 6 is fully inspected and without defects. Then the container 6 is ready to be transported away by the Container Transport System 7 and to be shipped to the customer. In the case of an error, the container 6 is removed from the CIT 2, the defect has to be corrected and checked again. All data is archived.

[0040] List of references

[0041] 1 User

[0042] 2 Container Inspection System 3 Rotatable Platform

[0043] 4 Computer / PLC

[0044] 5 Digital Camera System

[0045] 6 Container

[0046] 7 Container Transport System 8 Software

[0047] 9 Human User Interface

[0048] 10 Digital Images

[0049] 11 Network

Claims

Patent claims1 . An automated inspection system (2) for containers (6) of different size and weight, comprising of at least one digital camera system (5) which can be mounted in different positions, a rotatable platform (3) for positioning the containers (6) so they can be scanned by the at least one digital camera system (5), wherein either the at least one digital camera system (5) or the rotatable platform (3) or both are adjustable in height, a container transport system (7) which transports the containers (6) to the platform (3), a computer (4) which is connected to the at least one digital camera system (5) and which has access to a database and a software (8) performed by the computer (4) which analyzes digital images (10) taken by the at least one digital camera system (5) and determines the quality of the scanned container (6).

2. System according to claim 1 , characterized in that, either the digital camera system (5) is used as a reader for a machine- readable code, in particular a data matrix code, which is attached to the container (6), or the system (2) comprises of at least one additional Code reader device which can be mounted in different positions and is adjustable in height.

3. System according any of the previous claims, characterized in that, the computer (4) is a Programmable Logic Controller (PLC) with a touch panel as User Interface (9).

4. System according any of the previous claims, characterized in that, network (11 ) connects the computer (4) and the at least one digital camera (5).

5. System according to claim 4, characterized in that, the database is either stored on an internal memory or on an external memory which is connected to the computer (4) via the network (11 ).

6. System according any of the previous claims, characterized in that, the containers (6) are varying in size from 1 kg up to 400kg.

7. System according any of the previous claims, characterized in that, the at least one digital camera system (5) is attached to a robot which enables the at least one digital camera system (5) to be movable in all axes and to scan the container (6) from all sides..

8. Method for inspecting containers (6) of different size and weight using a system (2) according to claims 1 to 7, the following steps comprising:• Transporting a container (6) to the rotatable platform (3) via the transport system (7) and adjust the height of either the at least one digital camera system (5) or the platform (3) or both depending on size and form of the container (6);• Scanning the container (6) from all sides by taking digital images (10) via the at least one digital camera (5);• Sending the digital image data to the computer (4);• Determining the quality of the scanned container (6) by the software (8) on the computer (4) analyzing the digital image data; and• If the determined quality fails a given quality standard labeling it as defect by the software (8).

9. Method according to claim 8, characterized in that, the software (8) uses a neural network to analyze the quality of the containers (6) and to determine a good or bad container.

10. Method according to claims 7 to 9, characterized in that, the computer (4) creates a documentation of the container inspection comprising all relevant inspection data and stores the documentation in a data archive in the database.

Images