WO2022170702A1 - 模具的状态监测方法、装置、工控机、存储介质及系统 - Google Patents

模具的状态监测方法、装置、工控机、存储介质及系统 Download PDF

Info

Publication number
WO2022170702A1
WO2022170702A1 PCT/CN2021/097967 CN2021097967W WO2022170702A1 WO 2022170702 A1 WO2022170702 A1 WO 2022170702A1 CN 2021097967 W CN2021097967 W CN 2021097967W WO 2022170702 A1 WO2022170702 A1 WO 2022170702A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
region
area
image
mold
Prior art date
Application number
PCT/CN2021/097967
Other languages
English (en)
French (fr)
Inventor
吴俊耦
程鑫
张翔
吉守龙
徐必业
吴丰礼
Original Assignee
广东拓斯达科技股份有限公司
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 广东拓斯达科技股份有限公司 filed Critical 广东拓斯达科技股份有限公司
Publication of WO2022170702A1 publication Critical patent/WO2022170702A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G06T7/001—Industrial image inspection using an image reference approach
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20024—Filtering details
    • G06T2207/20032—Median filtering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • 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
    • Y02P90/30—Computing systems specially adapted for manufacturing

Definitions

  • the embodiments of the present application relate to the technical field of mold protection, for example, to a mold state monitoring method, device, industrial computer, storage medium, and system.
  • Condition monitoring of molds based on machine vision is an effective automatic mold monitoring solution.
  • This solution has strong versatility, but with the increase of the number of collected templates or the change of mold stations, the detection speed and detection speed are affected. Accuracy will have an impact, therefore, how to take into account the detection speed and detection accuracy has become the focus of research in mold condition monitoring.
  • the embodiments of the present application provide a mold state monitoring method, device, industrial computer, storage medium and system, which can take into account the accuracy and efficiency of mold state monitoring, and improve production efficiency while protecting the mold.
  • the embodiments of the present application provide a method for monitoring the state of a mold, including:
  • the template image is an image of the target mold in a normal state, and the template image includes a region of interest, and the region of interest includes the target mold Image;
  • a pyramid template matching algorithm is used to identify and locate the to-be-detected image to obtain the target area of the to-be-detected image;
  • the target mold is abnormal.
  • an embodiment of the present application provides a state monitoring device for a mold, including:
  • An image acquisition module configured to acquire a template image of the target mold and an image to be detected of the target mold;
  • the template image is an image of the target mold in a normal state, and the template image includes a region of interest, the the region contains the image of the target mold;
  • the image processing module is configured to identify and locate the to-be-detected image by using a pyramid template matching algorithm according to the region of interest of the template image, to obtain a target area of the to-be-detected image; according to the target area and the The pixel difference existing between the regions of interest determines the sensitive sub-region included in the target region; and according to the area of the sensitive sub-region, it is determined whether the target mold is abnormal.
  • an embodiment of the present application provides an industrial computer, including a memory, a processor, and a computer program stored in the memory and running on the processor.
  • the processor executes the computer program, the above-mentioned first The state monitoring method of the mold described in the aspect.
  • an embodiment of the present application provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the mold according to the first aspect above is implemented. Condition monitoring method.
  • an embodiment of the present application provides a state monitoring system for a mold, including:
  • the visual imaging component is set to collect the template image and the image to be detected of the object to be detected;
  • An industrial computer includes a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the method for monitoring the state of a mold as described in the first aspect when the processor executes the computer program.
  • FIG. 1 is a schematic flowchart of a state monitoring method for a mold provided in Embodiment 1 of the present application;
  • FIG. 2 is a schematic flowchart of an implementation of the state monitoring method for a mold provided in Embodiment 1 of the present application in an actual scene;
  • FIG. 3 is a schematic diagram of a pixel difference map provided in Embodiment 1 of the present application.
  • FIG. 4 is a schematic diagram of a sensitive sub-region provided in Embodiment 1 of the present application.
  • FIG. 5 is a schematic logical diagram of anomaly detection provided in Embodiment 1 of the present application.
  • FIG. 6 is a schematic flowchart of a preprocessing process of a template image provided in Embodiment 2 of the present application;
  • FIG. 7 is a schematic structural diagram of a state monitoring device for a mold provided in Embodiment 3 of the present application.
  • FIG. 8 is a schematic structural diagram of an industrial computer according to Embodiment 4 of the present application.
  • FIG. 9 is a schematic structural diagram of a mold state monitoring system according to Embodiment 6 of the present application.
  • the embodiment of the present application provides a technical solution for state monitoring of a mold.
  • the positioning and recognition speed of the target region is improved, thereby improving the For the detection speed of the mold state, by determining the sensitive sub-region, and detecting the abnormal state of the mold based on the area of the sensitive sub-region, the influence of interference factors is considered, and the detection accuracy of the mold state is improved.
  • FIG. 1 is a schematic flowchart of a state monitoring method for a mold provided in Embodiment 1 of the present application.
  • the method in this embodiment can be executed by the state monitoring device for a mold provided in an embodiment of the present application, and the device can be implemented by software and/or It can be realized by means of hardware or hardware, and can be integrated into the industrial computer.
  • the state monitoring method of the mold in this embodiment includes S101 to S104 .
  • the template image and the image to be detected of the same target mold collected by the visual imaging component are obtained, wherein the template image is the image of the target mold in a normal state, that is, when the visual imaging component is used
  • the image of the target mold is collected, and the template image can be processed in advance and stored according to the identification (such as number) of the target mold.
  • the corresponding template image can be extracted from the database storing the template image.
  • the image to be detected is an image used to determine whether the current state of the mold is abnormal, and is a real-time image of the target mold collected and uploaded by the visual imaging component.
  • the template image is processed in advance to determine the region of interest (ROI) of the template image.
  • the stencil image includes a region of interest that contains the image of the target stencil.
  • the region of interest in this embodiment is mainly the region where the image of the target mold is located.
  • the timing for state detection of the target mold can be set in advance, for example, after each mold opening or after the product is taken out, an image acquisition instruction is sent to the visual imaging component, so that the visual imaging component collects the target mold The to-be-detected image, thereby triggering the mold state detection process of the present application.
  • the template images used for state monitoring of the mold at different times can be different.
  • the template image used for state detection after the mold is opened (for the convenience of distinction, the template image is called the first template image) is different from the product.
  • the template image used for state detection (for the convenience of distinction, the template image is called the second template image)
  • the second template image After taking out the template image used for state detection (for the convenience of distinction, the template image is called the second template image), it is stored separately, and the state of the mold in different production stages is detected according to the first template image and the second template image, respectively. In this way, the state of the mold can be better monitored, the quality of the mold can be ensured, and the production process can be ensured without interruption, and the production efficiency can be improved.
  • FIG. 2 is a schematic diagram of an implementation flow of the mold state monitoring method provided in the first embodiment of the present application in an actual scene.
  • the monitor monitoring in FIG. 2 refers to determining whether the target mold is abnormal or not through the technical solution of this embodiment.
  • the technical solution of the present application can be applied in different stages such as after starting the detection, after the mold is opened, and after the product is taken out, so as to realize the state monitoring of the mold at each stage of product production, so as to better protect the product. mold.
  • a pyramid template matching algorithm is used to identify and locate the image to be detected, so as to obtain a target region of the image to be detected.
  • the pyramid template matching algorithm is used to identify and locate the image to be detected based on the region of interest of the template image of the target mold obtained in S101, and the target of the image to be detected is obtained. area.
  • the target area is an image area on the image to be detected that matches the area of interest of the template image, and correspondingly, the target area is also the area where the image of the target mold is located.
  • the pyramid template matching algorithm is an algorithm that superimposes the pyramid model and the template matching algorithm, which can reduce the amount of calculation.
  • the ROI is used as a sliding window, which starts to slide from the origin of the image to be detected, traverses the image to be detected, slides one step at a time, calculates the similarity between an ROI and the area covered by the ROI on the image to be detected, and determines the maximum value of the similarity , and determine the area corresponding to the maximum similarity value as the target area.
  • the pyramid algorithm is capable of creating images at multiple resolutions, the bottom of the pyramid is a high-resolution image of the image to be processed, and the top is a low-resolution image. Therefore, the higher the level of the pyramid, the lower the size and resolution. Accurate positioning can be performed through fast matching of low-resolution, layer-by-layer mapping, and positioning to the bottom high-resolution image.
  • the identification and positioning of the target area is the first step in the state detection of the target mold. After that, it is necessary to determine whether the target mold has abnormal state based on the target area. The slider is misplaced, the mold release is poor, etc. Accordingly, there will be pixel differences between the captured image and the image in the normal state. Therefore, in this step, the sensitive sub-areas contained in the target area are determined according to the differences between the target area and multiple pixels in the area of interest, so as to achieve the purpose of determining whether the target mold is abnormal.
  • a pixel difference map is generated based on the pixel difference between the target area and the area of interest, and the pixel difference map includes the pixel difference between the corresponding pixels in the target area and the area of interest value; according to the pixel difference value and the sensitivity threshold in the pixel difference map, determine the sensitive sub-areas contained in the target area.
  • FIG. 3 is a schematic diagram of the pixel difference map provided in Embodiment 1 of the present application, and the pixels of the region of interest and the target area are both 10 ⁇ 10 as an example.
  • each point on the pixel difference map is It is called a pixel difference point.
  • the value of each pixel difference point is the absolute value of the difference between the pixel value (such as gray value) of the corresponding pixel point in the target area and the area of interest, that is, the pixel difference. value.
  • the sensitivity threshold is a preset value used to determine whether the pixel difference between the target area corresponding to the pixel difference point and the area of interest requires special attention.
  • the sensitivity threshold value can be based on actual operating conditions or relevant experience. etc. to set, there is no restriction here.
  • the sensitive sub-region is the region formed by the pixel difference points whose pixel difference value is greater than the sensitivity threshold on the pixel difference map. point, called the target pixel difference point.
  • the target pixel difference point whose pixel difference value on the pixel difference map is greater than the sensitivity threshold is determined; and the sensitive sub-region included in the target area is determined according to the positional relationship between the target pixel difference points.
  • the positional relationship between the target pixel difference points refers to the adjacent relationship between the target pixel difference points.
  • the area formed by the target pixel difference points with adjacent relationships is determined as a sensitive sub-area. It can be seen that each pixel difference point in the sensitive sub-area is the target pixel difference point, that is, The pixel difference value of each pixel difference point in the sensitive sub-region is greater than the sensitivity threshold.
  • the sensitivity threshold is 20
  • the value of the pixel difference point on the pixel difference map whose pixel difference value is greater than 20 is replaced by 1
  • the value of the pixel difference point on the pixel difference map whose pixel difference value is less than or equal to 20 is replaced. is 0, obtain a pixel difference map represented by 1 and 0, and determine the sensitive sub-areas included in the target area according to the pixel difference points with a pixel difference value of 1 having an adjacent relationship, exemplarily, FIG.
  • the schematic diagram of the sensitive sub-region provided in Example 1 is shown in Figure 4.
  • the area filled with diagonal lines in the figure is the sensitive sub-region.
  • the area of at least one sensitive sub-region determined in S103 is determined, and according to the determined area of the sensitive sub-region, it is determined whether the target mold is abnormal, so as to achieve the purpose of detecting the state of the mold.
  • the area of the sensitive sub-region is the number of pixel difference points included in the sensitive sub-region.
  • the areas of the sensitive sub-regions P1, P2, P3, P4, and P5 in FIG. 4 are 5 , 12, 6, 3, and 1.
  • FIG. 5 is a logical schematic diagram of abnormality detection provided in Embodiment 1 of the application.
  • the target is first determined. Whether there is a sensitive sub-region with a single area greater than the first area threshold (the first area threshold is represented as S1 in FIG. 5 ) in the area; if there is a single area in the target area greater than the first area threshold, it is determined that the target mold is abnormal; In the case where there is no single area in the target area that is larger than the first area threshold, it is determined whether the total area of at least one sensitive sub-area is greater than the second area threshold (the second area threshold is represented as S2 in FIG. 5 ); When the total area of the regions is greater than the second area threshold, it is determined that the target mold is abnormal.
  • the first area threshold is represented as S1 in FIG. 5
  • the second area threshold is represented as S2 in FIG. 5
  • the number of sensitive sub-regions included in the determined target region is 0 through S103, it is directly determined that the target mold is not abnormal. If the number of sensitive sub-regions included in the target region determined in S103 is one, then in S104 it is only necessary to determine whether the area of the sensitive sub-region is greater than the first area threshold. If it is less than or equal to the first area threshold, it is determined that the target mold is normal; if the number of sensitive sub-areas contained in the target area determined in S103 is greater than 1, it needs to be judged strictly according to the judgment logic shown in FIG. 5 .
  • the target mold is determined to be normal; through this judgment mechanism, It can effectively ensure the accuracy of abnormal detection of the target mold.
  • the method of this embodiment further includes: if it is determined that the target mold is abnormal, generating an alarm.
  • an alarm Such as controlling the flickering of the human-computer interaction interface, controlling the alarm to give an alarm, controlling the sending of a prompt message to the user equipment (such as a mobile phone), etc. to alert the user to prompt the user to perform manual intervention in time, or directly control the injection molding machine to stop working, etc., so as to realize the target mold. protection of.
  • the template image is an image of the target mold in a normal state, the template image includes a region of interest, and the region of interest includes an image of the target mold;
  • the pyramid template matching algorithm is used to identify and locate the image to be detected to obtain the target region of the image to be detected; according to the pixel difference between the target region and the region of interest, determine the Sensitive sub-area: According to the area of the sensitive sub-area, it is determined whether the target mold is abnormal, which realizes automatic monitoring of mold quality, and takes into account the real-time and accuracy of mold abnormality detection. It can not only accurately monitor mold quality, but also improve Production efficiency has beneficial economic benefits and social value.
  • FIG. 6 is a schematic flowchart of the preprocessing process of the template image provided in the second embodiment of the application, as shown in FIG. 6 .
  • the preprocessing of the template image includes S201 to S203.
  • S201 Determine a pre-selected area of the template image according to a frame selection operation performed by a user on the target mold in the template image on the human-computer interaction interface.
  • the industrial computer provided by this embodiment includes a human-computer interaction interface, and the user can view the image collected by the visual imaging component through the human-computer interaction interface, and perform corresponding operations on the image.
  • the region of interest of the template image is determined to ensure the accuracy of the determined target region of the image to be detected.
  • the preselected area of the template image is determined by recognizing the frame selection operation performed by the user on the target mold in the template image on the human-computer interaction interface.
  • this embodiment in order to adapt to different shapes of molds, in this embodiment, users are supported to perform frame selection operations on template images in any shape, such as circles, rectangles, and polygons. At the same time, this embodiment also supports users to select multiple Pre-area to meet usage requirements in different scenarios, thereby enhancing the scenario adaptability of the technical solutions provided in this application.
  • a prompt message such as "please select the position of the mold”
  • a prompt message can be displayed on the human-computer interaction interface to guide the user's operation, thereby ensuring that the determined pre-selected area contains the image of the target mold.
  • selection icons of different shapes may also be set on the human-computer interaction interface for the user to select and use when performing a frame selection operation, thereby improving the convenience of the operation.
  • Boundary expansion is performed on the preselected area obtained in S201, so as to obtain the region of interest of the template image.
  • a fixed value filling method may be used to expand the selected area to a certain extent, so as to accurately identify the target area in the image to be detected subsequently.
  • the boundary of the preselected area is expanded to obtain the region of interest; If there are multiple pre-selected regions, based on the minimum circumscribed rectangle (or minimum circumscribed circle, etc.) of all the pre-selected regions, and according to the preset fill value, the boundary of the pre-selected region is expanded to obtain the region of interest.
  • the template image is subjected to adaptive median filtering processing.
  • an adaptive median filter may be set in advance, and the window size of the median filter can be dynamically changed by the adaptive median filter according to preset conditions, so as to realize the filtering processing of the image.
  • the adaptive median filter can not only filter out the salt and pepper noise with high probability, but also better protect the details of the image, which helps to improve the speed of subsequent image processing.
  • the image to be detected can also be subjected to adaptive median filtering processing as required.
  • the pre-selected area of the template image is determined according to the frame selection operation performed by the user on the target mold in the template image on the human-computer interaction interface, and the boundary of the pre-selected area is expanded to obtain the area of interest of the template image.
  • the adaptive median filtering process is performed to realize the preprocessing of the template image, determine the region of interest of the template image, and reduce unnecessary noise interference, which is beneficial to improve the speed and accuracy of subsequent image processing, thereby improving mold abnormality detection. the quality of.
  • FIG. 7 is a schematic structural diagram of a mold state monitoring device provided in Embodiment 3 of the present application. As shown in FIG. 7 , the mold state monitoring device 10 in this embodiment includes:
  • Image acquisition module 11 and image processing module 12 are identical to Image acquisition module 11 and image processing module 12 .
  • the image acquisition module 11 is configured to acquire a template image of the target mold and an image to be detected of the target mold; the template image is an image of the target mold in a normal state, and the template image includes a region of interest, a region of interest an image containing the target mold;
  • the image processing module 12 is configured to identify and locate the to-be-detected image by using a pyramid template matching algorithm according to the region of interest of the template image to obtain the target area of the to-be-detected image; According to the pixel difference existing between the regions of interest, the sensitive sub-region included in the target region is determined; according to the area of the sensitive sub-region, it is determined whether the target mold is abnormal.
  • the image processing module 12 is set to:
  • the pixel difference map includes pixel difference values of corresponding pixels in the target area and the interest area
  • Sensitive sub-regions included in the target region are determined according to the pixel difference value and the sensitivity threshold in the pixel difference map.
  • the image processing module 12 is set to:
  • the sensitive sub-areas included in the target area are determined, and each pixel difference point in the sensitive sub-area is a target pixel difference point.
  • the number of the sensitive sub-regions is at least one, and the image processing module 12 is set to:
  • determining whether there is a sensitive sub-region with a single area greater than a first area threshold in the target area in response to the presence of a single sensitive sub-region with a single area greater than the first area threshold in the target area, determine that the target mold is abnormal;
  • the image processing module 12 is also set to:
  • Boundary expansion is performed on the preselected region to obtain the region of interest of the template image.
  • the image processing module 12 is set to:
  • the boundary of the pre-selected area is expanded based on the minimum circumscribed rectangle of the pre-selected area to obtain the region of interest;
  • the preselected regions are expanded based on the minimum circumscribed rectangles of the multiple preselected regions to obtain the region of interest.
  • the image processing module 12 is also set to:
  • Adaptive median filtering is performed on the template image and the to-be-detected image.
  • the image processing module 12 is also set to:
  • the mold state monitoring device provided in this embodiment can execute the mold state monitoring method provided by the above method embodiments, and has functional modules corresponding to the execution method.
  • the implementation principles and technical effects of this embodiment are similar to those of the foregoing method embodiments, and details are not repeated here.
  • FIG. 8 is a schematic structural diagram of an industrial computer provided in Embodiment 4 of the application.
  • the industrial computer 20 includes a memory 21, a processor 22, and a computer program stored in the memory and running on the processor.
  • the number of the processor 22 in the industrial computer 20 can be at least one, and a processor 22 is taken as an example in FIG. 8; Connecting via a bus is an example.
  • the memory 21 can be configured to store software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the image acquisition module 11 and the image processing module 12 in the embodiments of the present application.
  • the processor 22 implements various functional applications and data processing of the industrial computer by running the software programs, instructions and modules stored in the memory 21 , that is, to implement the above-mentioned mold state monitoring method.
  • the memory 21 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like.
  • the memory 21 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device.
  • memory 21 may include memory located remotely from processor 22, which may be connected to the industrial computer through a grid. Examples of such grids include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
  • the fifth embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a computer processor, is used to execute a method for monitoring the state of a mold, and the method includes:
  • the template image is an image of the target mold in a normal state, the template image includes a region of interest, and the region of interest includes the image of the target mold ;
  • a pyramid template matching algorithm is used to identify and locate the to-be-detected image to obtain the target area of the to-be-detected image;
  • the target mold is abnormal.
  • the computer program of the computer-readable storage medium provided by the embodiment of the present application is not limited to the above-mentioned method operations, and can also perform the relevant operations in the state monitoring method of the mold provided by any embodiment of the present application. .
  • the present application can be implemented by means of software and necessary general-purpose hardware, and certainly can also be implemented by hardware.
  • the technical solutions of the present application can be embodied in the form of software products in essence or the parts that make contributions to related technologies, and the computer software products can be stored in a computer-readable storage medium, such as a computer floppy disk, Read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including several instructions to make a computer device (which can be a personal computer, A server, or a grid device, etc.) executes the methods described in the various embodiments of the present application.
  • a computer device which can be a personal computer, A server, or a grid device, etc.
  • FIG. 9 is a schematic structural diagram of the mold state monitoring system provided in the sixth embodiment of the application.
  • the mold state detection system 30 in this embodiment includes a visual imaging component 31 and the industrial computer 20 in the fourth embodiment. .
  • the visual imaging component 31 is configured to collect the template image and the to-be-detected image of the object to be detected.
  • the visual imaging component includes: an industrial camera, an imaging lens, an infrared light source, and an infrared filter.
  • an industrial camera an imaging lens, an infrared light source, and an infrared filter.
  • the following will describe the industrial camera, imaging lens, infrared light source, and infrared filter used in the embodiment of the present application. The situation is introduced:
  • industrial cameras with various resolutions such as 1 million pixel resolution, 3 million pixel resolution, and 5 million pixel resolution, can be provided, so as to facilitate the application in different application scenarios.
  • the detection object and accuracy requirements of the corresponding resolution are selected to improve the scene adaptability of the system.
  • the data transmission interface of the industrial camera in this embodiment is USB 3.0, which can not only perform data transmission, but also supply power to the industrial camera at the same time, without additional power supply, and has the advantages of easy use, fast transmission speed, and compatibility. good features.
  • the lens interface on the front of the camera is C-mount, which is a universal interface for easy lens replacement.
  • the chip of the industrial camera used in this embodiment is a black and white photosensitive chip.
  • the reasons for the selection are as follows: first, in the application scenario of mold monitoring, there is no need to have special requirements for color; secondly, due to cameras with the same resolution, black and white photosensitive chips are higher in accuracy than color photosensitive chips, especially for image edge detection, The black and white photosensitive chip has better imaging effect; finally, in the process of image processing, the black and white industrial camera obtains grayscale information, which can be processed directly.
  • variable zoom lens is selected in this system.
  • the focal length of the variable zoom lens in this embodiment is in the range of 12-50 mm.
  • infrared light source is used to illuminate, and an infrared filter is added to the front of the camera to ensure that infrared light enters the camera, so as to improve the anti-interference of the monitoring system to the environment.
  • the industrial computer 20 further includes:
  • Human-computer interaction interface used for information display and acquisition of operation instructions.
  • the human-computer interface includes the following functions:
  • Monitoring status display display the current working status in real time.
  • Production status display count the number of production inspections since the last reset.
  • Camera settings The user can set the camera exposure time, camera gain, product photo delay, product re-inspection delay, mold cavity photography delay and template re-inspection delay on the human-computer interface to achieve camera settings.
  • Parameter setting It is mainly the setting of detection parameters, including the number of detections, detection delay, administrator password setting, time setting and screen calibration, etc.
  • Template sampling The status bar in the upper left corner of the human-computer interaction interface will display the start of sampling, and the automatic sampling can be performed after the safety door is closed and the mold opening is completed.
  • Template setting Manually take pictures to select templates, draw monitoring range, add and delete templates and other functions.
  • Sensitivity Set the sensitivity of the detection area, area control and detection verification functions. Among them, the sensitivity and area can be set for a single area, which directly determines the range of defect detection. Inspection verification can bypass the signal and manually monitor the mold status.
  • Log record record the operation log and view normal/abnormal pictures, the main function is to view the alarm picture and alarm location, analyze the cause of the alarm, etc.
  • Signal display Display the current signal transmission and the next signal transmission.
  • the units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be realized; , the specific names of the functional units are only for the convenience of distinguishing from each other, and are not used to limit the protection scope of the present application.

Abstract

一种模具的状态监测方法、装置、工控机、存储介质及系统，该方法包括：获取目标模具的模板图像和目标模具的待检测图像，模板图像中包括感兴趣区域，感兴趣区域包含目标模具的图像（S101）；根据模板图像的感兴趣区域，采用金字塔模板匹配算法，对待检测图像进行识别定位，得到待检测图像的目标区域（S102）；根据目标区域与感兴趣区域之间存在的像素差异，确定目标区域中包含的敏感子区域（S103）；根据敏感子区域的面积，确定目标模具是否存在异常（S104）。

Description

模具的状态监测方法、装置、工控机、存储介质及系统
本申请要求在2021年2月9日提交中国专利局、申请号为202110178510.8的中国专利申请的优先权，该申请的全部内容通过引用结合在本申请中。
技术领域
本申请实施例涉及模具保护技术领域，例如涉及一种模具的状态监测方法、装置、工控机、存储介质及系统。
背景技术
近年来，汽车、建筑、家用电器、食品、医药等产业对注塑制品日益增长的需要，推动了整个注塑、成型技术水准的发展和提高，也带动了注塑加工业的发展。在注塑行业中，模具是重要的成型设备，其质量好坏直接关系到产品最终质量的好坏。由于模具在注塑加工生产成本中所占的比重较大，其使用寿命直接影响产品成本，为了保护模具，对模具的生产状态进行监视和检测是十分有必要的。
基于机器视觉对模具进行状态监测，是一种较为有效的自动化模具监测方案，该方案具备较强的通用性，但是随着采集模板数量的增加或者模具工位的变化，对检测速度和检测的精度会有影响，因此，如何兼顾检测速度和检测精度成为模具状态监测中研究的重点。
发明内容
本申请实施例提供一种模具的状态监测方法、装置、工控机、存储介质及系统，能够兼顾模具状态监测的准确性的高效性，在保护模具的同时，提高了生产效率。
第一方面，本申请实施例提供一种模具的状态监测方法，包括：
获取目标模具的模板图像和目标模具的待检测图像；所述模板图像为所述 目标模具在正常状态下的图像，所述模板图像中包括感兴趣区域，所述感兴趣区域包含所述目标模具的图像；
根据所述模板图像的感兴趣区域，采用金字塔模板匹配算法，对所述待检测图像进行识别定位，得到所述待检测图像的目标区域；
根据所述目标区域与所述感兴趣区域之间存在的像素差异，确定所述目标区域中包含的敏感子区域；
根据所述敏感子区域的面积，确定所述目标模具是否存在异常。
第二方面，本申请实施例提供一种模具的状态监测装置，包括：
图像获取模块，设置为获取目标模具的模板图像和目标模具的待检测图像；所述模板图像为所述目标模具在正常状态下的图像，所述模板图像中包括感兴趣区域，所述感兴趣区域包含所述目标模具的图像；
图像处理模块，设置为根据所述模板图像的感兴趣区域，采用金字塔模板匹配算法，对所述待检测图像进行识别定位，得到所述待检测图像的目标区域；根据所述目标区域与所述感兴趣区域之间存在的像素差异，确定所述目标区域中包含的敏感子区域；根据所述敏感子区域的面积，确定所述目标模具是否存在异常。
第三方面，本申请实施例提供一种工控机，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如上述第一方面所述的模具的状态监测方法。
第四方面，本申请实施例提供一种计算机可读存储介质，所述计算机可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现如上述第一方面所述的模具的状态监测方法。
第五方面，本申请实施例提供一种模具的状态监测系统，包括：
视觉成像组件，设置为采集待检测物品的模板图像和待检测图像；
工控机，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如上述第一方面所述的模具 的状态监测方法。
附图说明
图1为本申请实施例一提供的模具的状态监测方法的流程示意图；
图2为本申请实施例一提供的模具的状态监测方法在实际场景中的一种实施流程示意图；
图3为本申请实施例一提供的像素差分图的示意图；
图4为本申请实施例一提供的敏感子区域的示意图；
图5为本申请实施例一提供的异常检测的逻辑示意图；
图6为本申请实施例二提供的模板图像的预处理过程的流程示意图；
图7为本申请实施例三提供的模具的状态监测装置的结构示意图；
图8为本申请实施例四提供的一种工控机的结构示意图；
图9为本申请实施例六提供的模具的状态监测系统的结构示意图。
具体实施方式
下面结合附图和实施例对本申请作详细说明。
本申请实施例提供一种模具的状态监测的技术方案，通过确定模板图像的感兴趣区域，并采用金字塔模板匹配算法确定待检测图像的目标区域，提高了目标区域的定位识别速度，进而提高了对模具状态的检测速度，通过确定敏感子区域，并基于敏感子区域的面积进行模具的状态异常检测，考虑了干扰因素的影响，提高了对模具状态的检测精度。
实施例一
示例性地，图1为本申请实施例一提供的模具的状态监测方法的流程示意图，本实施例的方法可以由本申请实施例所提供的模具的状态监测装置执行，该装置可以由软件和/或硬件的方式来实现，并可集成于工控机中。如图1所示，本实施例的模具的状态监测方法，包括S101至S104。
S101、获取目标模具的模板图像和目标模具的待检测图像。
本步骤中，在需要对目标模具进行状态检测时，获取视觉成像组件采集的同一目标模具的模板图像和待检测图像，其中，模板图像为目标模具在正常状态下的图像，即在由视觉成像组件在目标模具质量完好且不存在残留物时，采集的目标模具的图像，模板图像可以事先处理好，并根据目标模具的标识(如编号)进行存储，在需要进行模具的状态监测时，从存储模板图像的数据库中提取相应的模板图像即可。待检测图像是用于确定模具当前是否存在状态异常的图像，是由视觉成像组件采集并上传的目标模具的实时图像。
需要说明的是，为提高待检测图像中目标模具的定位识别速度，本实施例中，事先对模板图像进行处理，确定模板图像的感兴趣区域(region of interest，ROI)，本步骤中，获取的模板图像中包括感兴趣区域，感兴趣区域包含目标模具的图像。可以理解的是，由于本实施例的主要目的是对模具进行状态监测，因此，本实施例中的感兴趣区域主要是目标模具的图像所在的区域。通过事先确定模板图像的感兴趣区域，并基于感兴趣区域进行目标区域的搜索，有利于提高图片的处理速度，进而提高目标模具状态的检测速度。
另外，本实施例中，可以事先设置对目标模具进行状态检测的时机，如在每次模具开模之后或产品取出之后，通过向视觉成像组件发送图像获取指令，以使视觉成像组件采集目标模具的待检测图像，从而触发本申请的模具状态检测流程。可以理解的是，在不同时刻对模具进行状态监测的模板图像可以不同，如通过将模具开模后进行状态检测使用的模板图像(为便于区分，将该模板图像叫做第一模板图像)与产品取出后进行状态检测使用的模板图像(为便于区分，将该模板图像叫做第二模板图像)进行分别存储，并根据第一模板图像和第二模板图像分别对不同生产阶段模具的状态进行检测，从而可以更好地对模具的状态进行监视，保证模具的质量，并保证生产流程不中断，提高生产效率。
示例性地，图2为本申请实施例一提供的模具的状态监测方法在实际场景中的一种实施流程示意图。图2中的监视器监视是指通过本实施例的技术方案， 确定目标模具是否存在异常。由图2可以看出，本申请的技术方案可应用在启动检测之后、模具开模之后和产品取出之后等不同的阶段，实现在产品生产的各个阶段对模具的状态监测，以更好地保护模具。
S102、根据模板图像的感兴趣区域，采用金字塔模板匹配算法，对待检测图像进行识别定位，得到待检测图像的目标区域。
在实际工况中，注塑机的老化导致模具开模位置不固定，会有些许偏差，导致在相机的成像面上存在偏差。因此，为了保证识别定位的准确性和速度，本步骤中，采用金字塔模板匹配算法，基于S101中获取的目标模具的模板图像的感兴趣区域，对待检测图像进行识别定位，得到待检测图像的目标区域。其中，目标区域是待检测图像上与模板图像的感兴趣区域相匹配的图像区域，相应地，目标区域也是目标模具的图像所在的区域。
其中，金字塔模板匹配算法是一种将金字塔模型和模板匹配算法叠加使用的一种算法，可以起到减少计算量的作用，示例性地，在识别定位目标区域时，本步骤中，以模板图像的ROI作为滑动窗口，从待检测图像的原点开始滑动，遍历待检测图像，每次滑动一个步长，计算一个ROI与待检测图像上被ROI覆盖的区域的相似度，确定相似度的最大值，并将相似度的最大值对应的区域确定为目标区域。由于金字塔算法能够创建多个分辨率的图像，金字塔的底部是待处理图像的高分辨率图像，而顶部是低分辨率的图像。因此，金字塔的层数越高，尺寸和分辨率就越低。通过低分辨率的快速匹配，逐层映射，一直定位到最底层的高分辨率图像，就可进行精确的定位。
S103、根据目标区域与感兴趣区域之间存在的像素差异，确定目标区域中包含的敏感子区域。
本实施例，进行目标区域的识别定位是对目标模具进行状态检测的第一步，之后，需要基于目标区域，确定目标模具是否存在状态异常，对模具来说，常见的状态异常包括物品残留、滑块错位、脱模不良等，相应地，采集到的图像与正常状态下的图像之间就会存在像素差异。因此，本步骤中，根据目标区域 与感兴趣区域的多个像素点之间存在的差异，确定目标区域中包含的敏感子区域，从而达到确定目标模具是否存在异常的目的。
在一种可能的实现方式中，本步骤中，基于目标区域与感兴趣区域之间存在的像素差异，生成像素差分图，像素差分图中包括目标区域与感兴区域中对应像素点的像素差异值；根据像素差分图中的像素差异值和敏感度阈值，确定目标区域中包含的敏感子区域。
示例性地，图3为本申请实施例一提供的像素差分图的示意图，以感兴趣区域和目标区域像素均为10×10为例，如图3所示，像素差分图上的每一个点叫做一个像素差分点，在像素差分图上，每个像素差分点的数值即为目标区域与感兴趣区域上对应像素点的像素值(如灰度值)的差值的绝对值，即像素差异值。
其中，敏感度阈值是事先设定的一个数值，用于确定该像素差分点对应的目标区域与感兴趣区域的像素差异是否需要格外关注，该敏感度阈值的数值可根据实际工况或相关经验等进行设定，此处不做限制。
可以理解的，由于外界环境、相机性能等因素，导致对同一物体在不同的时刻采集的图片本身可能存在一些像素差异，与模具的状态无关，并且这些像素差异往往在一个较小的范围内，因此，本实施例中，通过设置敏感度阈值以将这些因素造成的影响去除掉。相应地，敏感子区域是像素差分图上像素差异值大于敏感度阈值的像素差分点构成的区域，为便于区分，本实施例中，将素差分图上像素差异值大于敏感度阈值的像素差分点，叫做目标像素差分点。
在一种可能的实现方式中，通过确定出像素差分图上像素差异值大于敏感度阈值的目标像素差分点；根据目标像素差分点之间的位置关系，确定目标区域中包含的敏感子区域。
其中，目标像素差分点之间的位置关系，是指目标像素差分点之间相邻关系。相应地，本步骤中，将具有相邻关系的目标像素差分点构成的区域，确定为一个敏感子区域，由此可见，敏感子区域中的每个像素差分点均为目标像素 差分点，即敏感子区域中每个像素差分点的像素差异值均大于敏感度阈值。
示例性地，假设敏感度阈值为20，将像素差分图上像素差异值大于20的像素差分点的数值替换为1，将像素差分图上像素差异值小于或等于20的像素差分点的数值替换为0，得到用1和0表示的像素差分图，并根据像素差异值为1的具有相邻关系的像素差分点确定目标区域中包括的敏感子区域，示例性地，图4为本申请实施例一提供的敏感子区域的示意图，如图4所示，图中斜线填充的区域即为敏感子区域，图4中的敏感子区域有5个，分别记为P1、P2、P3、P4和P5。
S104、根据敏感子区域的面积，确定目标模具是否存在异常。
本步骤中，确定S103中确定的至少一个敏感子区域的面积，并根据确定的敏感子区域的面积，确定目标模具是否存在异常，从而达到检测模具状态的目的。
示例性地，本实施例中，敏感子区域的面积为敏感子区域包括的像素差分点的个数，相应地，图4中敏感子区域P1、P2、P3、P4和P5的面积分别为5、12、6、3和1。
在一种可能的实现方式中，图5为本申请实施例一提供的异常检测的逻辑示意图，本步骤中通过图5所示的判断机制确定目标模具是否存在异常，示例性地，先确定目标区域中是否存在单个面积大于第一面积阈值(图5中第一面积阈值表示为S1)的敏感子区域；在目标区域中存在单个面积大于第一面积阈值的情况下，确定目标模具存在异常；在目标区域中不存在单个面积大于第一面积阈值的情况下，则确定至少一个敏感子区域的总面积是否大于第二面积阈值(图5第二面积阈值表示为S2)；在至少一个敏感子区域的总面积大于第二面积阈值的情况下，确定目标模具存在异常。
示例性地，若预先设置第一面积阈值为10，第二面积阈值为30，由于P2的面积为12，大于第一面积阈值，因此，可直接确定目标模具存在异常，不需要再进行总面积检测。若预先设置的第一面积阈值为15，第二面积阈值为25， 先进行单个面积检测，确定P1、P2、P3、P4和P5的面积均小于第一面积阈值，示例性地，进行总面积检测，发现P1、P2、P3、P4和P5的总面积为5﹢12﹢6﹢3﹢1＝27，大于第二面积阈值，从而确定目标模具存在异常。
需要说明的是，若通过S103，确定的目标区域中包含的敏感子区域的个数为0个时，则直接确定目标模具不存在异常。若S103中确定的目标区域中包含的敏感子区域的个数为1个时，则S104中只需要判断该敏感子区域的面积是否大于第一面积阈值即可，若判断该敏感子区域的面积小于或等于第一面积阈值即可，则确定目标模具正常；若S103中确定的目标区域中包含的敏感子区域的个数大于1个时，则需要严格按照图5所示的判断逻辑进行判断，只有当每个敏感子区域的面积均小于或等于第一面积阈值，且多个敏感子区域的面积之和小于或等于第二面积阈值时，才确定目标模具正常；通过这种判断机制，可有效保证对目标模具进行异常检测的精度。
可选地，在S104之后，本实施例的方法还包括：若确定目标模具存在异常，则进行告警。如控制人机交互界面闪烁、控制报警器报警、控制向用户设备(如手机)发送提示消息等进行告警，以提示用户及时进行人工干预，或者直接控制注塑机停止工作等，从而实现对目标模具的保护。
本实施例中，通过获取目标模具的模板图像和目标模具的待检测图像，模板图像为目标模具在正常状态下的图像，模板图像中包括感兴趣区域，感兴趣区域包含目标模具的图像；根据模板图像的感兴趣区域，采用金字塔模板匹配算法，对待检测图像进行识别定位，得到待检测图像的目标区域；根据目标区域与所述感兴趣区域之间存在的像素差异，确定目标区域中包含的敏感子区域；根据敏感子区域的面积，确定目标模具是否存在异常，实现了对模具质量的自动化监测，并且兼顾了模具异常检测的实时性和准确性，不仅能够准确监视模具质量，还提高了生产效率，具有有益的经济效益和社会价值。
实施例二
下面将通过一个具体的实施例对目标模具的模板图像的预处理过程加以说明，示例性地，图6为本申请实施例二提供的模板图像的预处理过程的流程示意图，如图6所示，本实施例中，对模板图像的预处理包括S201至S203。
S201、根据用户在人机交互界面对模板图像中目标模具进行的框选操作，确定模板图像的预选区域。
本实施例提供的工控机包括人机交互界面，用户可以通过人机交互界面查看视觉成像组件采集的图像，并对图像进行相应的操作。在一种可能的实现方式中，本实施例中，在用户对模板图像进行的框选操作的基础上，确定模板图像的感兴趣区域，以保证确定的待检测图像的的目标区域的准确性。示例性地，本步骤中，通过识别用户在人机交互界面对模板图像中目标模具进行的框选操作，确定模板图像的预选区域。
需要说明的是，为适应模具的不同形状，本实施例中支持用户对模板图像进行任意形状的框选操作，如圆形、矩形和多边形等，同时，本实施例中也支持用户选择多个预先区域，以满足不同场景下的使用需求，从而增强了本申请提供的技术方案的场景适应能力。
可以理解的是，本实施例中，可以通过在人机交互界面显示提示消息，如“请框选出模具的位置”，以引导用户的操作，从而保证确定的预选区域中包含目标模具的图像。另外，本实施例中，还可以在人机交互界面上设置不同形状的选择图标，以供用户进行框选操作时选择使用，提高操作的便捷性。
S202、对预选区域进行边界扩充，得到模板图像的感兴趣区域。
由于注塑机老化或机械波动等原因，模具或视觉成像组件的位置可能会出现一些细微的移动，从而使不同的图片之间可能会存在些许位置偏差，本步骤中，为了更好的获取边缘，对S201中得到的预选区域进行边界扩充，从而得到模板图像的感兴趣区域。
可选地，本步骤中，可以采用固定值填充的方式，将选定区域进行一定程度的扩大，以便后续对待检测图像中的目标区域进行准确的识别定。
在一种可能的实现方式中，若预选区域为一个，则基于该预选区域的最小外接矩形(或最小外接圆等)，根据预设填充值，对预选区域进行边界扩充，得到感兴趣区域；若预选区域有多个，则基于所有预选区域的最小外接矩形(或最小外接圆等)，根据预设填充值，对预选区域进行边界扩充，得到感兴趣区域。
S203、对模板图像进行自适应中值滤波处理。
本步骤中，为更好地保护模板图像中的相关细节，在完成感兴趣区域的确定以后，对模板图像进行自适应中值滤波处理。
可选地，实施例中，可以事先设置自适应中值滤波器，通过自适应中值滤波器根据预设好的条件，动态地改变中值滤波器的窗口尺寸，实现对图像的滤波处理。自适应中值滤波器不但能够滤除概率较大的椒盐噪声，而且能够更好的保护图像的细节，有助于提高后续图像处理的速度。
可以理解的是，在对待检测图像进行定位识别之前，根据需要，也可以对待检测图像进行自适应中值滤波处理。
本实施例中，通过根据用户在人机交互界面对模板图像中目标模具进行的框选操作，确定模板图像的预选区域，对预选区域进行边界扩充，得到模板图像的感兴趣区域，对模板图像进行自适应中值滤波处理，实现了对模板图像预处理，确定了模板图像的感兴趣区域，并减少了不必要的噪声干扰，有利于提高后续图像处理的速度和精度，进而提高模具异常检测的质量。
实施例三
图7为本申请实施例三提供的模具的状态监测装置的结构示意图，如图7所示，本实施例中模具的状态监测装置10包括：
图像获取模块11和图像处理模块12。
图像获取模块11，设置为获取目标模具的模板图像和目标模具的待检测图像；所述模板图像为所述目标模具在正常状态下的图像，所述模板图像中包括感兴趣区域，感兴趣区域包含所述目标模具的图像；
图像处理模块12，设置为根据所述模板图像的感兴趣区域，采用金字塔模板匹配算法，对所述待检测图像进行识别定位，得到所述待检测图像的目标区域；根据所述目标区域与所述感兴趣区域之间存在的像素差异，确定所述目标区域中包含的敏感子区域；根据所述敏感子区域的面积，确定所述目标模具是否存在异常。
可选地，图像处理模块12是设置为：
基于所述目标区域与所述感兴趣区域之间存在的像素差异，生成像素差分图，所述像素差分图中包括所述目标区域与所述感兴区域中对应像素点的像素差异值；
根据所述像素差分图中的像素差异值和敏感度阈值，确定所述目标区域中包含的敏感子区域。
可选地，图像处理模块12是设置为：
确定出所述像素差分图上像素差异值大于所述敏感度阈值的目标像素差分点；
根据目标像素差分点之间的位置关系，确定所述目标区域中包含的敏感子区域，所述敏感子区域中的每个像素差分点均为目标像素差分点。
可选地，所述敏感子区域的个数为至少一个，图像处理模块12是设置为：
确定所述目标区域中是否存在单个面积大于第一面积阈值的敏感子区域；响应于所述目标区域中存在单个面积大于第一面积阈值的敏感子区域，确定所述目标模具存在异常；
响应于所述至少一个敏感子区域的总面积大于第二面积阈值，确定至少一个敏感子区域的总面积是否大于第二面积阈值；响应于所述至少一个敏感子区域的总面积大于第二面积阈值，确定所述目标模具存在异常。
可选地，图像处理模块12还设置为：
根据用户在人机交互界面对所述模板图像中所述目标模具进行的框选操作，确定所述模板图像的预选区域；
对所述预选区域进行边界扩充，得到所述模板图像的感兴趣区域。
可选地，图像处理模块12是设置为：
若所述预选区域为一个，则基于该预选区域的最小外接矩形对所述预选区域进行边界扩充，得到所述感兴趣区域；
若所述预选区域有多个，则基于多个预选区域的最小外接矩形对所述预选区域进行边界扩充，得到所述感兴趣区域。
可选地，图像处理模块12还设置为：
对所述模板图像和所述待检测图像进行自适应中值滤波处理。
可选地，图像处理模块12还设置为：
若确定所述目标模具存在异常，则进行告警。
本实施例所提供的模具的状态监测装置可执行上述方法实施例所提供的模具的状态监测方法，具备执行方法相应的功能模块。本实施例的实现原理和技术效果与上述方法实施例类似，此处不再一一赘述。
实施例四
图8为本申请实施例四提供的一种工控机的结构示意图，如图8所示，该工控机20包括存储器21、处理器22及存储在存储器上并可在处理器上运行的计算机程序；工控机20中的处理器22的数量可以是至少一个，图8中以一个处理器22为例；工控机20中的处理器22、存储器21可以通过总线或其他方式连接，图8中以通过总线连接为例。
存储器21作为一种计算机可读存储介质，可设置为存储软件程序、计算机可执行程序以及模块，如本申请实施例中的图像获取模块11和图像处理模块12对应的程序指令/模块。处理器22通过运行存储在存储器21中的软件程序、指令以及模块，从而实现工控机的各种功能应用以及数据处理，即实现上述的模具的状态监测方法。
存储器21可主要包括存储程序区和存储数据区，其中，存储程序区可存储 操作系统、至少一个功能所需的应用程序；存储数据区可存储根据终端的使用所创建的数据等。此外，存储器21可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。在一些实例中，存储器21可包括相对于处理器22远程设置的存储器，这些远程存储器可以通过网格连接至工控机。上述网格的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
实施例五
本申请实施例五还提供一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序在由计算机处理器执行时用于执行一种模具的状态监测方法，该方法包括：
获取目标模具的模板图像和目标模具的待检测图像；所述模板图像为所述目标模具在正常状态下的图像，所述模板图像中包括感兴趣区域，感兴趣区域包含所述目标模具的图像；
根据所述模板图像的感兴趣区域，采用金字塔模板匹配算法，对所述待检测图像进行识别定位，得到所述待检测图像的目标区域；
根据所述目标区域与所述感兴趣区域之间存在的像素差异，确定所述目标区域中包含的敏感子区域；
根据所述敏感子区域的面积，确定所述目标模具是否存在异常。
当然，本申请实施例所提供的一种包计算机可读存储介质，其计算机程序不限于如上所述的方法操作，还可以执行本申请任意实施例所提供的模具的状态监测方法中的相关操作。
通过以上关于实施方式的描述，所属领域的技术人员可以清楚地了解到，本申请可借助软件及必需的通用硬件来实现，当然也可以通过硬件实现。基于这样的理解，本申请的技术方案本质上或者说对相关技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品可以存储在计算机可读存储介 质中，如计算机的软盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、闪存(FLASH)、硬盘或光盘等，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网格设备等)执行本申请各个实施例所述的方法。
实施例六
图9为本申请实施例六提供的模具的状态监测系统的结构示意图，如图9所示，本实施例中的模具的状态检测系统30包括视觉成像组件31和实施例四中的工控机20。
其中，视觉成像组件31，设置为采集待检测物品的模板图像和待检测图像。
可选地，本实施例中，视觉成像组件包括：工业相机、成像镜头、红外光源和红外滤光片，下面将对本申请实施例使用的工业相机、成像镜头、红外光源和红外滤光片的情况加以介绍：
(1)工业相机
为了更好的应对应用场景，本实施例中，可提供多种分辨率的工业相机，如100万像素分辨率、300万像素分辨率和500万像素分辨率等，以便于针对不同应用场景下的检测对象和精度要求选择对应分辨率的相机，从而提高本系统的场景适应性。
示例性地，本实施例中的工业相机的数据传输接口为USB 3.0，该接口不仅可进行数据传输，还可同时为工业相机供电，无需额外的电源，具有使用简便，传输速度快，兼容性好等特点。相机前端的镜头接口为C口，作为通用接口，方便镜头替换。
可选地，本实施例中使用的工业相机的芯片为黑白感光芯片。选择理由如下：首先，在模具监视的应用场景中，不需要对颜色有特别要求；其次，由于同等分辨率的相机，在精度上，黑白感光芯片比彩色感光芯片高，尤其针对图像边缘检测，黑白感光芯片成像效果更佳；最后，在图像处理过程中，黑白工 业相机得到的是灰度信息，可直接处理。
(2)成像镜头
由于镜头直接决定视野大小和清晰程度，镜头选型是否合适直接决定视觉成像模块能否采集成像质量较高的图像。在实际的应用场景中，由于不同产品和模具的特殊性、不规则性和尺寸因素，不适用定焦镜头。因此，在该系统选用可变焦镜头，示例性地，本实施例中可变焦镜头的焦距大小范围为12-50mm，配合光圈和清晰度的调整，能够使得检测对象处于合适的大小和亮度。
(3)红外光源和红外滤光片
为避免外界光照对后续图像处理的影响，采用红外光源照射，并且在相机前端添加红外滤光片以保证红外光进入相机，以提高监视系统的对环境的抗干扰性。
可选地，工控机20还包括：
人机交互界面，用于进行信息展示和操作指令的获取。示例性地，人机交互界面包括如下功能：
监控状态显示：实时显示当前工作状态。
生产状态显示：统计上次清零后生产检测的数量。
相机设置：用户可在人机交互界面设置相机曝光时间、相机增益、产品拍照延时、产品复检延时、模腔拍照延时和模板复检延时等，以实现对相机的设置。
参数设置：主要为检测参数的设置，包括检测次数、检测延时、管理员密码设置、时间设置和屏幕校准等。
模板采样：人机交互界面的左上角状态栏会显示开始采样，等待安全门关闭和开模完信号，可进行自动采样。
模板设置：进行手动拍照选取模板，绘制监控范围，模板数量添加和删除等功能。
灵敏度：设置检测区域的灵敏度、面积控制和检测验证功能。其中，灵敏 度和面积可针对单个区域设置，直接决定缺陷检测的范围。检测验证可绕开信号，手动监测模具状态。
日志记录：记录操作日志和查看正常/异常图片，主要作用是查看报警图片和报警位置，分析报警原因等。
信号显示：显示当前信号传输情况和下一步信号传输。
值得注意的是，上述模具的状态监测装置的实施例中，所包括的各个单元和模块只是按照功能逻辑进行划分的，但并不局限于上述的划分，只要能够实现相应的功能即可；另外，各功能单元的具体名称也只是为了便于相互区分，并不用于限制本申请的保护范围。

Claims (16)

  1. 一种模具的状态监测方法，包括：
    获取目标模具的模板图像和目标模具的待检测图像；所述模板图像为所述目标模具在正常状态下的图像，所述模板图像中包括感兴趣区域，所述感兴趣区域包含所述目标模具的图像；
    根据所述模板图像的感兴趣区域，采用金字塔模板匹配算法，对所述待检测图像进行识别定位，得到所述待检测图像的目标区域；
    根据所述目标区域与所述感兴趣区域之间存在的像素差异，确定所述目标区域中包含的敏感子区域；
    根据所述敏感子区域的面积，确定所述目标模具是否存在异常。
  2. 根据权利要求1所述的方法，其中，所述根据所述目标区域与所述感兴趣区域之间存在的像素差异，确定所述目标区域中包含的敏感子区域，包括：
    基于所述目标区域与所述感兴趣区域之间存在的像素差异，生成像素差分图，所述像素差分图中包括所述目标区域与所述感兴区域中对应像素点的像素差异值；
    根据所述像素差分图中的像素差异值和敏感度阈值，确定所述目标区域中包含的敏感子区域。
  3. 根据权利要求2所述的方法，其中，所述根据所述像素差分图中的像素差异值和敏感度阈值，确定所述目标区域中包含的敏感子区域，包括：
    确定出所述像素差分图上像素差异值大于所述敏感度阈值的目标像素差分点；
    根据目标像素差分点之间的位置关系，确定所述目标区域中包含的敏感子区域，所述敏感子区域中的每个像素差分点均为目标像素差分点。
  4. 根据权利要求1所述的方法，其中，所述敏感子区域的个数为至少一个，所述根据所述敏感子区域的面积，确定所述目标模具是否存在异常，包括：
    确定所述目标区域中是否存在单个面积大于第一面积阈值的敏感子区域；响应于所述目标区域中存在单个面积大于第一面积阈值的敏感子区域，确定所 述目标模具存在异常；
    响应于所述目标区域中不存在单个面积大于第一面积阈值的敏感子区域，确定至少一个敏感子区域的总面积是否大于第二面积阈值；响应于所述至少一个敏感子区域的总面积大于第二面积阈值，确定所述目标模具存在异常。
  5. 根据权利要求1-4任一项所述的方法，所述获取目标模具的模板图像和待检测图像之前，还包括：
    根据用户在人机交互界面对所述模板图像中所述目标模具进行的框选操作，确定所述模板图像的预选区域；
    对所述预选区域进行边界扩充，得到所述模板图像的感兴趣区域。
  6. 根据权利要求5所述的方法，其中，所述对所述预选区域进行边界扩充，得到所述模板图像的感兴趣区域，包括：
    在所述预选区域为一个的情况下，基于所述预选区域的最小外接矩形对所述预选区域进行边界扩充，得到所述感兴趣区域；
    在所述预选区域有多个的情况下，基于多个预选区域的最小外接矩形对所述预选区域进行边界扩充，得到所述感兴趣区域。
  7. 根据权利要求1-4任一项所述的方法，所述根据所述模板图像的感兴趣区域，采用金字塔模板匹配算法，对所述待检测图像进行识别定位，得到所述待检测图像的目标区域之前，还包括：
    对所述模板图像和所述待检测图像进行自适应中值滤波处理。
  8. 根据权利要求1-4任一项所述的方法，还包括：
    响应于确定所述目标模具存在异常，进行告警。
  9. 一种模具的状态监测装置，包括：
    图像获取模块，设置为获取目标模具的模板图像和目标模具的待检测图像；所述模板图像为所述目标模具在正常状态下的图像，所述模板图像中包括感兴趣区域，所述感兴趣区域包含所述目标模具的图像；
    图像处理模块，设置为根据所述模板图像的感兴趣区域，采用金字塔模板 匹配算法，对所述待检测图像进行识别定位，得到所述待检测图像的目标区域；根据所述目标区域与所述感兴趣区域之间存在的像素差异，确定所述目标区域中包含的敏感子区域；根据所述敏感子区域的面积，确定所述目标模具是否存在异常。
  10. 一种工控机，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如权利要求1-8中任一所述的模具的状态监测方法。
  11. 一种计算机可读存储介质，所述计算机可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现如权利要求1-8中任一所述的模具的状态监测方法。
  12. 一种模具的状态监测系统，包括：
    视觉成像组件，设置为采集待检测物品的模板图像和待检测图像；
    工控机，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如权利要求1-8中任一所述的模具的状态监测方法。
  13. 根据权利要求12所述的系统，其中，所述视觉成像组件包括：工业相机、成像镜头、红外光源和红外滤光片。
  14. 根据权利要求13所述的系统，其中，所述工业相机采用的芯片为黑白感光芯片。
  15. 根据权利要求13所述的系统，其中，所述成像镜头为可变焦镜头。
  16. 根据权利要求12所述的系统，其中，所述工控机还包括：
    人机交互界面，用于进行信息展示和操作指令的获取。
PCT/CN2021/097967 2021-02-09 2021-06-02 模具的状态监测方法、装置、工控机、存储介质及系统 WO2022170702A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110178510.8 2021-02-09
CN202110178510.8A CN112837302B (zh) 2021-02-09 2021-02-09 模具的状态监测方法、装置、工控机、存储介质及系统

Publications (1)

Publication Number Publication Date
WO2022170702A1 true WO2022170702A1 (zh) 2022-08-18

Family

ID=75933238

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/097967 WO2022170702A1 (zh) 2021-02-09 2021-06-02 模具的状态监测方法、装置、工控机、存储介质及系统

Country Status (2)

Country Link
CN (1) CN112837302B (zh)
WO (1) WO2022170702A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116309602A (zh) * 2023-05-24 2023-06-23 济南章力机械有限公司 基于机器视觉的数控钻铣床工作状态检测方法

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112837302B (zh) * 2021-02-09 2024-02-13 广东拓斯达科技股份有限公司 模具的状态监测方法、装置、工控机、存储介质及系统
CN113469974B (zh) * 2021-07-05 2022-12-02 天津市三特电子有限公司 球团链篦机篦板状态监测方法及其监测系统
CN113592831B (zh) * 2021-08-05 2024-03-19 北京方正印捷数码技术有限公司 印刷误差的检测方法、装置和存储介质
CN114299026A (zh) * 2021-12-29 2022-04-08 广东利元亨智能装备股份有限公司 检测方法、装置、电子设备及可读存储介质
CN115049713B (zh) * 2022-08-11 2022-11-25 武汉中导光电设备有限公司 图像配准方法、装置、设备及可读存储介质
CN115953397B (zh) * 2023-03-13 2023-06-02 山东金帝精密机械科技股份有限公司 一种圆锥轴承保持器的工艺制备流程的监测方法及设备
CN116935079B (zh) * 2023-09-07 2024-02-20 深圳金三立视频科技股份有限公司 一种基于视觉的线状开关状态监测方法及终端
CN117095000B (zh) * 2023-10-19 2024-01-26 杭州和利时自动化有限公司 一种设备检测方法及装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9349076B1 (en) * 2013-12-20 2016-05-24 Amazon Technologies, Inc. Template-based target object detection in an image
CN107336417A (zh) * 2017-06-13 2017-11-10 上海斐讯数据通信技术有限公司 一种基于机器视觉的模具保护方法及系统
CN110426395A (zh) * 2019-07-02 2019-11-08 广州大学 一种太阳能el电池硅片表面检测方法及装置
CN111474184A (zh) * 2020-04-17 2020-07-31 河海大学常州校区 基于工业机器视觉的aoi字符缺陷检测方法和装置
CN112837302A (zh) * 2021-02-09 2021-05-25 广东拓斯达科技股份有限公司 模具的状态监测方法、装置、工控机、存储介质及系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9349076B1 (en) * 2013-12-20 2016-05-24 Amazon Technologies, Inc. Template-based target object detection in an image
CN107336417A (zh) * 2017-06-13 2017-11-10 上海斐讯数据通信技术有限公司 一种基于机器视觉的模具保护方法及系统
CN110426395A (zh) * 2019-07-02 2019-11-08 广州大学 一种太阳能el电池硅片表面检测方法及装置
CN111474184A (zh) * 2020-04-17 2020-07-31 河海大学常州校区 基于工业机器视觉的aoi字符缺陷检测方法和装置
CN112837302A (zh) * 2021-02-09 2021-05-25 广东拓斯达科技股份有限公司 模具的状态监测方法、装置、工控机、存储介质及系统

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116309602A (zh) * 2023-05-24 2023-06-23 济南章力机械有限公司 基于机器视觉的数控钻铣床工作状态检测方法
CN116309602B (zh) * 2023-05-24 2023-08-04 济南章力机械有限公司 基于机器视觉的数控钻铣床工作状态检测方法

Also Published As

Publication number Publication date
CN112837302A (zh) 2021-05-25
CN112837302B (zh) 2024-02-13

Similar Documents

Publication Publication Date Title
WO2022170702A1 (zh) 模具的状态监测方法、装置、工控机、存储介质及系统
US11430103B2 (en) Method for image processing, non-transitory computer readable storage medium, and electronic device
US9773302B2 (en) Three-dimensional object model tagging
WO2021042816A1 (zh) 一种检测监控设备故障的方法、装置
WO2019227762A1 (zh) 光学模组像素缺陷检测方法、装置及设备
JP4416795B2 (ja) 補正方法
US20170261319A1 (en) Building height calculation method, device, and storage medium
US20130021477A1 (en) Method and camera for determining an image adjustment parameter
CN106604005B (zh) 一种投影电视自动对焦方法及系统
KR20220058843A (ko) 카메라 렌즈의 상태 검출 방법, 장치, 기기 및 저장 매체
CN110248101B (zh) 对焦方法和装置、电子设备、计算机可读存储介质
WO2017206444A1 (zh) 一种成像差异检测方法、装置及计算机存储介质
JP2014057303A (ja) 深度推定手順において改善されたシーン検出を利用するためのシステム及び方法
CN111368819B (zh) 光斑检测方法和装置
CN104052933A (zh) 动态范围模式的判定方法及其图像获取装置
CN110049240B (zh) 摄像头控制方法、装置、电子设备和计算机可读存储介质
WO2016101524A1 (zh) 纠正被摄物体拍摄倾斜方法、装置及移动终端、存储介质
CN104463827A (zh) 一种图像采集模块的自动检测方法及相应的电子设备
CN110365897B (zh) 图像修正方法和装置、电子设备、计算机可读存储介质
US10965858B2 (en) Image processing apparatus, control method thereof, and non-transitory computer-readable storage medium for detecting moving object in captured image
CN109068060B (zh) 图像处理方法和装置、终端设备、计算机可读存储介质
CN110780982A (zh) 一种图像处理方法、装置及设备
CN110650288B (zh) 对焦控制方法和装置、电子设备、计算机可读存储介质
CN110399823B (zh) 主体跟踪方法和装置、电子设备、计算机可读存储介质
CN108737733B (zh) 信息提示方法和装置、电子设备、计算机可读存储介质

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: 21925356

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE