WO2012125526A1 - Mécanisme de suppression d'éléments indésirables uniques pour une utilisation avec une vision artificielle sur une ligne de production - Google Patents

Mécanisme de suppression d'éléments indésirables uniques pour une utilisation avec une vision artificielle sur une ligne de production Download PDF

Info

Publication number
WO2012125526A1
WO2012125526A1 PCT/US2012/028709 US2012028709W WO2012125526A1 WO 2012125526 A1 WO2012125526 A1 WO 2012125526A1 US 2012028709 W US2012028709 W US 2012028709W WO 2012125526 A1 WO2012125526 A1 WO 2012125526A1
Authority
WO
WIPO (PCT)
Prior art keywords
reject
camera
signal
machine vision
vision system
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/US2012/028709
Other languages
English (en)
Inventor
Wei-Song TAN
David KOSEN
Thomas R. NEWTON
Brian K. RIDDLE
Stephen R. WACHTEL
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ACCUSENTRY Inc
Original Assignee
ACCUSENTRY Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by ACCUSENTRY Inc filed Critical ACCUSENTRY Inc
Publication of WO2012125526A1 publication Critical patent/WO2012125526A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Program-control systems
    • G05B19/02—Program-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G05B19/41815—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by the cooperation between machine tools, manipulators and conveyor or other workpiece supply system, workcell
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T1/00—General purpose image data processing
    • G06T1/0007—Image acquisition
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/34—Director, elements to supervisory
    • G05B2219/34388—Detect correct moment, position, advanced, delayed, then next command
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30164—Workpiece; Machine component
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Definitions

  • single cull on a high-speed manufacturing production line comprises an improvement over current inspection systems, especially machine vision inspection systems. It is a common practice in high-speed manufacturing to inspect a product with a machine vision inspection system and physically reject a product from the production line when a defective product is detected. However, because of certain uncertainties inherent in the propagation of a reject signal between the inspection system and the reject gate, it is a common practice to reject three or more products when only one defective product is detected.
  • the system and method according to the present invention will facilitate a single cull of all defective products only and is comprised of one or more CCD cameras, an image processing computer with hardware and an adaptive phasing module means for addressing problems encountered in prior art systems. The uncertainty or non-deterministic aspects of the reject signal propagation can be avoided by processing the camera reject signal based on current conditions.
  • Figure 1 illustrates the timing requirement of the cull signal at the reject gate.
  • the cull signal must be issued at the precise time, taking into account the delay of the cull action response time due to such characteristics as the reject gate mechanism, in order to catch the defective product while it moves through the reject gate at high speed.
  • the timing for the image capture is controlled by the trigger signal in response to the production line system programmable logic controller (PLC), which is input to the camera.
  • PLC production line system programmable logic controller
  • the required phasing of the trigger signal is dictated by the position of the camera and the required timing to capture a complete view of the product.
  • the reference time of the trigger signal is in the abstract is independent of the Reject Gate Reference Time, which is a production line system parameter and not a characteristic of the camera or of the machine vision system used on the production line.
  • the trigger signal is referenced to the Reject Gate Reference Time for specific applications. The difference between the two Reference Times is dependent upon the mounting position of the camera along the travel direction of the web.
  • the relative position between the reference time of the Reject Gate and the Camera Trigger is a function of the distance between the camera and the reject gate and the length of the product.
  • the Reference Time Difference is the distance between the camera station and the reject gate subtracted from the closest multiples of the product length and divided by the velocity or speed of the production line. Since product lengths vary depending on product styles produced on the same production line, it is impractical to control the distance of the camera from the Reject Gate. Furthermore, the position of the camera is determined by the available space at the production line where certain components must be inspected. Yet, as will be shown below, the Reference Time Difference between the Reject Gate and the Camera Trigger is very critical to the success of achieving single culling.
  • Figure 3 shows the Reject Signal Shift Mechanism for synchronizing the signal from the camera station to the reject gate.
  • the Reference Time Difference between the reject gate and the camera trigger must not exceed a certain maximum value.
  • the maximum value is determined by a combination of the timing requirement for the shift register operation and the minimum cycle time.
  • the minimum cycle time is dictated by the fastest line speed while running the smallest product.
  • Figure 3 appears to show the camera reject signal at the same time as the camera trigger signal. However, the camera reject signal shown in Figure 3 is actually the reject signal for a previous camera image, which is saved to be output from the machine vision system at a constant time.
  • the simplest arrangement is that the Camera Reject Signal is sent from the machine vision system as soon as the image processing is complete. This is inferior to the typical arrangement which is employed in the current invention and in other machine vision systems where the Camera Reject Signal is at least held until the next Camera Trigger is received. This is important because at least the Camera Reject Signal will be output at a consistent time. In the simplest arrangement, the Camera Reject Signal could be output at any time in the product cycle.
  • the phasing of the Camera Trigger Reference Time must be adjusted to allow the camera to capture a full product view. As illustrated in Figure 4, the "Reference Time difference between the reject gate and Camera Trigger" plus the required "Setup Time” must be less than the product cycle time.
  • the Setup Time is the minimum amount of time the reject signal must be ready before the Reject PLC shift register clock event in order for the signal to be reliably, or deterministically, sampled.
  • the Hold Time is the minimum amount of time the reject signal must be valid after the Reject PLC shift register clock event in order for the signal to be deterministically sampled.
  • a machine vision system is used for rejecting only individual defective products manufactured on a continuous production line, including production line electronic control apparatus, without rejecting proximate product that are not defective as a safeguard against failure to reject a defective product.
  • the machine vision system includes an input for receiving cyclical reject gate reference time signals that comprise a parameter of the production line.
  • a camera which is part of the machine vision system, receives a camera trigger signal for activating the camera. The camera trigger signal is referenced to the cyclical reject gate reference time signals.
  • An output from the machine vision system sends a camera reject signal to the production line electronic control apparatus during a second reject reference time cycle subsequent to a first reject time cycle in which the camera is activated for activation of a single reject gate to cull only one individual product at a time without culling the proximate products that are not defective.
  • the camera reject signal is delayed relative to the camera trigger signal in addition to the delay between the first and second reference time cycles to avoid a non-deterministic time zone proximate to receipt of a new reject gate reference time signal indicating a new reject gate cycle, so that the reject signal is sent during a time period in which a reject gate can be activated only during the passage of a single defective product.
  • This invention includes a machine vision system for inspecting product on a continuous production line to identify defective products and to cull all defective products at a reject gate on the production line without the need to cull proximate, but non-defective products, in order to insure a complete cull process.
  • the machine vision system has at least one camera.
  • An input to the machine vision system receives cyclical reject gate reference time or position signals that comprise a parameter of the production line to generate a camera trigger signal in response to the cyclical reject gate reference time or position signals.
  • a processor in the machine vision system generates a reject signal for a defective product in response to examination of an image from the camera, the reject signal being sent to the reject gate to cull a defective product.
  • the reject signal is delayed by a constant time or position delay relative to the cyclical reject gate reference time or position signals and to the camera trigger signal to avoid sending the reject signal during a non-deterministic time interval in which a non-defective product may be in the reject gate.
  • Figure 1 shows Cull Signal Timing at the Reject Gate.
  • Figure 2 shows the Reject Timing Signal at the Camera Station.
  • Figure 3 shows the Reject Signal Shift Mechanism.
  • Figure 4 shows the Camera Trigger Timing Constraint as it relates to a Non-Deterministic Period.
  • Figure 5 shows the manner in which the Camera can be repositioned according to the Prior Art.
  • Figure 6 shows how an Automatic Delay of the Camera Reject Signal can avoid the non-deterministic zone.
  • Figure 7 shows the prior art manner in which an Inspection System, such as a Machine Vision System, can be integrated with production line components that would not be part of the Machine Vision System.
  • Inspection System such as a Machine Vision System
  • Figure 8 shows the manner in which an Inspection System, such as a Machine Vision System according to this invention, can be integrated with production line components.
  • Inspection System such as a Machine Vision System according to this invention
  • Figure 9 is a diagram of a first embodiment of Machine Vision Inspection System according to this invention, which includes an Adaptive Phasing Module.
  • Figure 10 shows the required input and output of a Machine Vision Inspection System according to a second embodiment of this invention to achieve Single Culling capability.
  • Figure 1 1 is a block diagram of the Machine Vision Inspection System according to the second embodiment of Figure 1 0 and the manner in which this Machine Vision System interacts with a Production Line Reject PLC.
  • Figure 12 shows a timing chart of the system shown in Figures 10 and 1 1 with one camera.
  • Figure 13 shows a timing chart of the system shown in Figures 10 and 1 1 with four cameras.
  • the embodiments of this invention introduce a delay in sending a reject signal instead of repositioning machine vision cameras on a high speed production line. Practically the position of the camera cannot be mechanically adjusted to match the signal timing requirements.
  • the timing of the reject signal must be automatically delayed to the next product cycle (See Cycle Time in Figure 4) relative to the Reject Gate Reference Time in order to ensure a reliable cull. This delay is demonstrated in Figure 6.
  • the number of required shifts for the Camera Reject Signal must be reduced by one to insure that he camera reject signal is transmitted to the reject PLC at the point where only the defective product will be aligned in the reject gate.
  • each camera is triggered at a certain time before the product reaches a product reject gate.
  • the reject signal is then entered into a shift register so that the gate can be activated when the defective product reaches the reject gate.
  • the "shift register” is a table inside the PLC which stores the pass/fail status of each product going down the assembly line. As each product approaches the reject gate, the shift register "shifts" based on but not necessarily the same as changes in the reject gate reference time cycles, so that each product is moved to the next row in the table. This is the means for tracking each product.
  • the underlying problem to be solved is that since the Reject PLC and the machine vision system are asynchronous, it is possible that when the camera reject signal for product A is sent from the machine vision system, the Reject PLC enters that result into its shift register under product B. Since all input and output signals are discrete signals only, the timing alone dictates which product corresponds to a signal.
  • the principle of the adaptive phasing module according to a first embodiment of this invention is to estimate the point in time at which the Reject PLC is shifting the shift register from Product A to Product B, and then delay our reject signal to a later time if needed. If a camera reject signal is output just as the Reject PLC shift register is shifting, then it is uncertain as to whether the reject will be attributed to Product A or B.
  • an Adaptive Phasing Module is introduced in-between the Main Production Line PLC and the Production Line Reject PLC.
  • the Reject PLC will generate the reject gate timing signal referenced earlier and send it to the adaptive phasing module, and in turn to the inspection system.
  • the inspection system will then use the reject gate timing signal to trigger each camera. Since each camera can be positioned at any point on the production line, it is necessary that each camera can be triggered by the inspection system at any time during the product cycle.
  • the PLS may be configured to derive a more consistent signal that will comprise and can be referred to as a Reject Gate Reference Time Signal for the purposes of this invention.
  • the inspection system will phase, or delay, the activation/trigger of each camera based on input from an encoder signal from the production line system so that camera images are synchronized with the product motion as required.
  • the encoder signal tells the machine vision system the product position. Specifically, the encoder is mounted on a shaft of the production line and sends a signal back to the machine vision system. The encoder sends a fixed number of pulses for each rotation of the shaft. For a line scan camera, each pulse will trigger the line scan camera to capture one line of the image.
  • the PLC gives a frame trigger to tell the cameras when the beginning of a product is in front of the camera, but the encoder generates the line triggers which activate the camera to capture each line of the image. But the encoder signal can also be used to simply track the movement of the product by distance rather than time. This is a more accurate measurement because the product does may not be traveling at a constant speed.
  • each camera trigger signal must be set a priori during the initial setup of the inspection system in order that each camera is taking a picture of one complete product. If a defective product is detected, the camera reject signal will be adaptively phased based on the reject gate timing signal and known information such as the Reject PLC required Setup and Hold Times. The inspection system will then send a single reject signal to the Reject PLC by way of the Adaptive Phasing Module at a time dictated by the specified number of cycles to account for the travel of a rejected product from the camera inspection location to the rejection gate plus an additional delay to avoid a nondeterministic zone, thus bypassing the Main PLC. By avoiding dependence on the PLS and the Main PLC, two elements of uncertainty are avoided. By automatically phasing the inspection system reject signal to avoid the non-deterministic zone of the Reject PLC, the last significant element of uncertainty has been avoided.
  • FIG. 9 describes the Adaptive Phasing Technology in more detail.
  • the Adaptive Phasing Module receives the Main Production Line PLC reject Signal, as well as an "enable" input.
  • This Main Production Line PLC reject signal refers to reject signals initiated by systems other than the Machine Vision System of this invention, which can be used with other reject signal generating systems or mechanisms.
  • the enable input allows the Main PLC to disable the Adaptive Phasing Module from sending reject signals to the Reject PLC.
  • the Adaptive Phasing Module also takes the reject gate timing signal from the Reject PLC and the encoder signal and sends them to the inspection system.
  • the inspection system is a Machine Vision System, such as the Sentry 9000 system from AccuSentry using the SmartVisionSystem network and described in more detail in US Patent Application 13/066,172 filed on April 8, 201 1 , incorporated herein by reference.
  • the encoder signal and reject gate timing signal are sent to the SVP camera controller of the SVM machine vision image processing computer.
  • the camera controller contains a field-programmable gate array (FPGA) which will compute the phasing of the camera triggers and control the phasing of the inspection system reject signal, as well as the Main PLC reject signal, based on system parameters, such as the position of the camera or cameras and the reject gate.
  • FPGA field-programmable gate array
  • the camera controller interfaces with the Sentry 9000 software and the SVP-IO Input-Output controller.
  • the Input-Output Controller sends and receives discrete input and output signals.
  • area scan and line scan cameras can be interchangeably daisy-chained together and require a single camera controller card.
  • the product images from each camera are transferred to the machine vision image processing computer to be processed to determine the pass or reject stated of each product.
  • the image processing computer may be running image processing software, such as Sentry 9000, from AccuSentry, Inc. Sentry 9000 allows the user to interface with the hardware to configure the various image capture
  • Sentry 9000 contains a graphical user interface (GUI) and a platform for developing complex image processing algorithms using a progression of pre-defined tools for detection and measurement of various image elements. Additionally, Sentry 9000 is capable of communicating with various other factory devices using a variety of communication protocols such as RS-232, TCP/IP, Ethernet/IP, OPC, RSLinx, and Modbus. Sentry 9000 is also capable of building a Human-Machine Interface (HMI) to allow manufacturing technicians and operators to safely and efficiently interface with the machine vision system. Typically, the image processing computer will be outfitted with a touch-screen display.
  • GUI graphical user interface
  • a second embodiment of this invention eliminates the separate Adaptive Phasing Module with an alternate version that is both more reliable and easier to set-up.
  • the "Adaptive Phasing Module” will not be a separate device, but rather its functionality will be included on the Machine Vision camera controller.
  • the cameras will not be triggered by the Production Line PLS, but rather, all cameras will be triggered based on the reject gate timing signal.
  • the cameras triggers will be generated by the camera controllers and will be independently phased within the machine vision system using the encoder input.
  • This second embodiment can also be used in the inspection of a mass produced product such as a disposable diaper.
  • a typical baby diaper is constructed from numerous components. Some components inside the finished product, and therefore are only observable at an earlier point of manufacture. For this reason, in order to inspect that every component is correct, is usually necessary to install multiple cameras at different points of the production line. By placing cameras earlier in the manufacturing process, when certain defects are created, they can be corrected sooner resulting in fewer defective products. In other situations, a particular defect is indicative of a continuous problem. By inspecting for this defect as early in the production line as possible, this defect can be detected as soon as possible. In certain situations excessive unplanned downtime can be avoided by pro- actively shutting the production line down before this problem propagates down the production line to cause a much bigger problem such as material jams or web breaks.
  • the camera reject signal in this second embodiment will no longer be adaptively phased based on the PLC setup and hold time requirements. Instead the camera reject signal will be phased such that it is always output 180 degrees out of phase from the camera trigger signal.
  • ancillary capabilities can also be added to the system.
  • the single cull system will be capable of acting as a PLS to send output signals at any point in the cycle time.
  • Camera reject signal could be delayed by position alone, rather than by product.
  • the system can be configured to reject two products when a defect is detected on the border between two discrete products, and it cannot be determine with certainty that the second product is not also defective.
  • the single camera trigger signal input into the Machine Vision System for each camera, and a single camera reject signal output from the Machine Vision System, as shown in Figure 10.
  • the camera trigger signal will be the reject gate reference time signal from the Reject PLC, which is a parameter of the Production Line System that will be used with this new Machine Vision System.
  • This Machine Vision System will also have input from an encoder on the production line. This will be the same encoder used by the line scan cameras to generate line triggers.
  • This Machine Vision System itself will generate a frame trigger for each camera at the appropriate time, based on pre-determined positional delays from the reject gate reference time signal. After processing each camera's image, this Machine Vision System will generate a single pass or fail camera reject signal.
  • This one camera reject signal will be output to the reject PLC. However, the camera reject signal will be delayed by 180°, which (for typical absorbent hygiene manufacturing speeds) will safely avoid the non-deterministic timing zone.
  • the reject PLC may now check the status of this camera reject signal and enter the pass or fail state into its shift register without risk of erroneous accounting.
  • the machine vision system may also contain a shift register with a configurable number of positions. The purpose of this shift register is to allow the machine vision system to buffer multiple images. Firstly, this allows the image acquisition to be performed while an earlier image is simultaneously processed. Secondly, the processing time of a particular image may exceed the product cycle time without necessarily causing an overrun state.
  • FIG. 1 is a diagram of the second system and method.
  • the SVP camera controller communicates with the cameras and the encoder.
  • the Reject Gate Reference Time signal and the Camera Reject Signal are input from and output to the Reject PLC.
  • An FPGA on the SVP camera controller will control the timing of the camera frame triggers and Camera Reject Signal.
  • the rising edge of the Reject Gate Reference Time indicates the beginning of each product. From this time, the machine vision system can generate a trigger signal for Camera 0 at the correct positional delay. After the machine vision system processes the Camera 0 image and determines the pass or reject result, this result is held in a shift register inside the machine vision system.
  • the shift register contains three positions.
  • the Camera Reject Signal is therefore delayed by three product cycles plus 1 80° before sending. In this manner, the output is sent at a time when the Reject PLC can deterministically register the pass or fail signal to the correct product.
  • the advantages of this arrangement include the following.
  • the system can reliably reject a single product, even when critical system parameters such as the camera trigger phasing are adjusted.
  • the system must be wired for both a camera trigger signal and a camera reject signal for each camera. Therefore a four camera system would require 8 signals, whereas use of the proposed method and system would require only 2 signals.
  • camera reject signals can be output with an entirely position based (as opposed to product based) delay.
  • the system can generate position based output signals for any purpose, replicating the function a PLS. This capability is valuable when a production line does not have an easily adjustable PLS, or does not have any free channels on the existing PLS.
  • this system Although the purpose of this system is to facilitate the culling of a single product, certain situations might dictate that multiple products should be rejected. For example, when a certain kind of defect is detected at the end of one product, it might stand to reason that the beginning of the next product will also exhibit the same defect. Therefore, if the inspection of the camera images for either product detects this type of defect, the system can send reject signals for both products even if no defects were detected in the corresponding other image. For example, when the system is used to inspect a product, such as a diaper that is assembled from multiple components, a material splice for one component might span across two products.
  • the inspection system detects a material splice at either the beginning or end of a product, it would be safest to also reject the corresponding leading or following product. If the material splice was detected in the center portion of a product and was not close to either end, then rejection of a single product is sufficient to ensure that the material splice is not sent to an end-customer.
  • the Main PLC might control the reject gate, and would therefore also serve as the Reject PLC.
  • the Reference Gate Reference Time signal may be provided from the PLS rather than the Reject PLC.
  • the Camera Reject Signal might be delayed by more or less than 180° to account for a special situation, such as a very high-speed application where the sum of the PLC setup time and hold time comprises a significant portion of the overall product cycle time. For example, if the PLC hold and setup times comprise more than 1 0% of the product cycle time.
  • the system could also send a separate Camera Reject Signals for each camera. Additionally, each the system could send multiple output signals for each camera to indicate the reject reason or an overrun state.
  • This Machine Vision System may communicate with the Main PLC, Reject PLC or any other factory device to communicate a variety of information. For example, production date, defect counters, product styles, enable/disable of any or all inspections, heartbeat, camera overrun, etc. may be communicated. This data may be communicated by Discrete IO signals, RS-232, TCP/IP, OPC, EthernetIP, Profinet, Modbus, etc.
  • Another common machine vision system is the "smart camera” arrangement.
  • the only component is the camera itself, which also controls the IO and processes the images.
  • this system consists of a single camera, but in other regards it would have the same challenges for single culling as the system we describe in the patent and the same approach to avoiding nondeterministic zones can be applied to a smart camera arrangement.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Quality & Reliability (AREA)
  • Automation & Control Theory (AREA)
  • Theoretical Computer Science (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
  • Image Processing (AREA)

Abstract

Un système de vision artificielle pour une utilisation dans un système de production à grande vitesse, tel que la fabrication de couches jetables, permet de ne rejeter que des produits défectueux uniques en tant que partie d'une opération de suppression d'éléments indésirables uniques. Pour obtenir une capacité de suppression d'éléments indésirables uniques, une zone non déterministe dans laquelle un signal de rejet provenant d'un système de vision artificielle relativement asynchrone est appliqué à un système de commande et d'activation de porte de rejet est évitée. Ce signal de rejet de système de vision artificielle est également retardé par rapport à un signal de déclenchement de caméra d'un temps ou d'une position en plus de retards de registre à décalage dictés par les paramètres du système.
PCT/US2012/028709 2011-03-11 2012-03-12 Mécanisme de suppression d'éléments indésirables uniques pour une utilisation avec une vision artificielle sur une ligne de production Ceased WO2012125526A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US201161464928P 2011-03-11 2011-03-11
US61/464,928 2011-03-11

Publications (1)

Publication Number Publication Date
WO2012125526A1 true WO2012125526A1 (fr) 2012-09-20

Family

ID=46831070

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2012/028709 Ceased WO2012125526A1 (fr) 2011-03-11 2012-03-12 Mécanisme de suppression d'éléments indésirables uniques pour une utilisation avec une vision artificielle sur une ligne de production

Country Status (1)

Country Link
WO (1) WO2012125526A1 (fr)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8712574B2 (en) 2009-06-02 2014-04-29 The Procter & Gamble Company Systems and methods for controlling phasing of advancing substrates in absorbent article converting lines
US8712573B2 (en) 2009-06-02 2014-04-29 The Procter & Gamble Company Systems and methods for controlling registration of advancing substrates in absorbent article converting lines
US8712570B2 (en) 2009-06-02 2014-04-29 The Procter & Gamble Company Systems and methods for detecting and rejecting defective absorbent articles from a converting line
WO2024055796A1 (fr) * 2022-09-14 2024-03-21 杭州海康机器人股份有限公司 Système, procédé et appareil de détection de défaut, dispositif électronique, et support de stockage
CN119327750A (zh) * 2024-10-28 2025-01-21 北京领视智联科技有限公司 一种空白包装袋缺陷检测及剔除方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6473169B1 (en) * 2000-05-03 2002-10-29 Air Logic Power Systems, Inc. Integrated leak and vision inspection system
US20050275834A1 (en) * 2004-06-09 2005-12-15 Silver William M Method and apparatus for locating objects
US20080182008A1 (en) * 2007-01-31 2008-07-31 Snow Gerald F Apparatus and method for coating and inspecting objects
US20100092032A1 (en) * 2008-10-10 2010-04-15 Remus Boca Methods and apparatus to facilitate operations in image based systems

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6473169B1 (en) * 2000-05-03 2002-10-29 Air Logic Power Systems, Inc. Integrated leak and vision inspection system
US20050275834A1 (en) * 2004-06-09 2005-12-15 Silver William M Method and apparatus for locating objects
US20080182008A1 (en) * 2007-01-31 2008-07-31 Snow Gerald F Apparatus and method for coating and inspecting objects
US20100092032A1 (en) * 2008-10-10 2010-04-15 Remus Boca Methods and apparatus to facilitate operations in image based systems

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8712574B2 (en) 2009-06-02 2014-04-29 The Procter & Gamble Company Systems and methods for controlling phasing of advancing substrates in absorbent article converting lines
US8712573B2 (en) 2009-06-02 2014-04-29 The Procter & Gamble Company Systems and methods for controlling registration of advancing substrates in absorbent article converting lines
US8712570B2 (en) 2009-06-02 2014-04-29 The Procter & Gamble Company Systems and methods for detecting and rejecting defective absorbent articles from a converting line
US9429937B2 (en) 2009-06-02 2016-08-30 The Procter & Gamble Company Systems and methods for detecting and rejecting defective absorbent articles from a converting line
US9429929B2 (en) 2009-06-02 2016-08-30 The Procter & Gamble Company Systems and methods for controlling phasing of advancing substrates in absorbent article converting lines
US10331113B2 (en) 2009-06-02 2019-06-25 The Procter & Gamble Company Systems and methods for detecting and rejecting defective absorbent articles from a converting line
US10327959B2 (en) 2009-06-02 2019-06-25 The Procter & Gamble Company Systems and methods for controlling phasing of advancing substrates in absorbent article converting lines
US10966877B2 (en) 2009-06-02 2021-04-06 The Procter & Gamble Company Method for assembling products in a manufacturing process
US11079746B2 (en) 2009-06-02 2021-08-03 The Procter & Gamble Company Method for inspecting products in a manufacturing process
WO2024055796A1 (fr) * 2022-09-14 2024-03-21 杭州海康机器人股份有限公司 Système, procédé et appareil de détection de défaut, dispositif électronique, et support de stockage
CN119327750A (zh) * 2024-10-28 2025-01-21 北京领视智联科技有限公司 一种空白包装袋缺陷检测及剔除方法

Similar Documents

Publication Publication Date Title
JP6190003B2 (ja) マルチフェーズ同期化を有する分散視覚システム
WO2012125526A1 (fr) Mécanisme de suppression d'éléments indésirables uniques pour une utilisation avec une vision artificielle sur une ligne de production
US9200890B2 (en) Machine vision systems and methods with predictive motion control
US9632492B2 (en) Redundant watchdog method and system utilizing safety partner controller
EP2903005A1 (fr) Système d'enregistrement d'événements vidéo
JP4935369B2 (ja) サーボシステム
CN108778637B (zh) 致动器控制系统、致动器控制方法及记录介质
CN104338684A (zh) 送料器用速度检测装置以及送料器
EP3223093A1 (fr) Dispositif de mesure optique
JP4078065B2 (ja) 複数のユニットで進行するプロセスを同期化する装置および方法
CN102176650A (zh) 主从伺服系统同步信号产生方法、装置以及灯检机
US8792566B2 (en) Method and apparatus for digital data transmission
CN118158023B (zh) 一种总线模块信号采集控制方法、装置、设备及存储介质
TWI739916B (zh) 同步化多個處理系統
CN117160923A (zh) 不受传送速度波动影响的不合格产品剔除闭环控制方法
US9880525B2 (en) Positioning apparatus and PLC system using same
US20220143818A1 (en) Counter unit, counter unit control method, control device, and control system
US6813664B2 (en) System transmitting data in equidistance cycles using successive synchronization signals for detecting and signaling access violations
US20160169709A1 (en) Device and method for checking a clock signal of a position measuring device
JP2001331346A (ja) シミュレータ及びシミュレーション方法
JP6523226B2 (ja) 時刻同期制御システム
JP2012173016A (ja) 検査システムの制御方法
JP6818659B2 (ja) クロック信号検査装置、プラント監視制御装置、およびクロック信号検査装置の診断方法
JPH04151791A (ja) 部品計数装置
JP4460181B2 (ja) 撮像システムおよびそれに用いる撮像装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12757374

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 12757374

Country of ref document: EP

Kind code of ref document: A1