Resin layering interface identification method, device and equipment

Abstract

The application provides a resin layering interface identification method, device and equipment. The method receives an image identification instruction, collects image information of a resin to be identified based on the image identification instruction, selects a region of interest in the image information based on image segmentation, divides the region of interest into preset equal parts in a top-to-bottom manner, determines a single-channel mean value of a preset number of lines at the bottom of each equal part, counts the number of single-channel mean values greater than a layering threshold value, and determines that a layering interface is identified when the number is greater than a preset value. The preset value is less than the preset number of lines. The technical solution of the application effectively reduces the amount of calculation by judging whether the resin is negative resin in the equal part, thereby effectively solving the freezing problem of the processing terminal and effectively improving the efficiency of resin layering identification.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image recognition, in particular to a resin layering interface recognition method, device and equipment. BACKGROUND

[0002] When the anion and cation resins of the high-speed mixed bed in the polishing system are regenerated in vitro, the anion and cation resins need to be separated, and resin layering often needs to be monitored and confirmed by manual on-site inspection. At present, most of the resin layering is confirmed by template matching based on gray value or image model comparison based on neural network self-learning training.

[0003] However, the template matching or model comparison method has a large amount of computer operation, is prone to terminal running lag, and has low resin layering recognition efficiency. SUMMARY

[0004] The present application provides a resin layering interface recognition method, device and equipment to solve the problem of low resin layering recognition efficiency in the prior art.

[0005] The present application provides a resin layering interface recognition method, which comprises:

[0006] receiving an image recognition instruction, and collecting image information of the resin to be recognized based on the image recognition instruction;

[0007] selecting a region of interest in the image information based on image segmentation;

[0008] dividing the region of interest into a preset number of equal parts from top to bottom;

[0009] determining the single-channel mean value of a preset number of rows at the bottom of each equal part;

[0010] counting the number of single-channel mean values greater than the layering threshold value;

[0011] when the number is greater than a preset value, it is determined that the layering interface is recognized, and the preset value is less than the preset number of rows.

[0012] According to the resin layering interface recognition method provided by the present application, after selecting the region of interest in the image information based on image segmentation, the method further comprises:

[0013] selecting a target region image in the region of interest;

[0014] obtaining the single-channel mean value of the target region image as the cation resin feature value;

[0015] adding a bias to the cation resin feature value to obtain an anion resin threshold value, and taking the anion resin threshold value as the layering threshold value.

[0016] The resin layer interface identification method provided by the application adds a bias to the positive resin characteristic value to obtain a negative resin threshold value, and comprises the following steps:

[0017] Determine the resin properties of different mixed beds and the resin properties of different pollution levels.

[0018] Determine the bias size based on the resin properties of different mixed beds and the resin properties of different pollution levels.

[0019] Add the bias size to the positive resin characteristic value to obtain a negative resin threshold value.

[0020] The resin layer interface identification method provided by the application receives an image identification instruction, acquires image information of the resin to be identified based on the image identification instruction, and comprises the following steps:

[0021] Receive a control instruction from a distributed control system through a programmable logic controller.

[0022] Control the camera to acquire image information of the resin to be identified by using a resin tank periscope based on the control instruction.

[0023] The resin layer interface identification method provided by the application further comprises the following steps:

[0024] When at least one of the following conditions is met, i.e., the acquisition signal of the camera is not identified, the control signal of the programmable logic controller is not identified, a device power failure signal is identified, and a calculation parameter overrun signal is identified, disconnect the signal transmission between the programmable logic controller and the distributed control system.

[0025] Issue an alarm prompt through the distributed control system.

[0026] The programmable logic controller and the distributed control system are connected through a hardwire in the resin layer interface identification method provided by the application.

[0027] The resin layer interface identification method provided by the application further comprises the following steps:

[0028] Identify the signal processing type in the resin layer interface identification process.

[0029] If the signal processing type is signal reading and threshold operation, use a main thread to process data.

[0030] If the signal processing type is function operation and video display, use a first sub-thread to process data.

[0031] If the signal processing type is a feedback layer signal, use a second sub-thread to process data.

[0032] The resin layer interface recognition method provided by the present application is based on image segmentation, and a region of interest is selected in image information, comprising:

[0033] The image information is divided into a plurality of mutually non-overlapping regions;

[0034] Pixel features of each region are extracted;

[0035] When resin features are recognized in the pixel features, the corresponding region is taken as the region of interest.

[0036] The present application also provides a resin layer interface recognition device, comprising:

[0037] A collection module is configured to receive an image recognition instruction, and collect image information of a resin to be recognized based on the image recognition instruction;

[0038] A segmentation module is configured to select a region of interest in the image information based on image segmentation;

[0039] A division module is configured to divide the region of interest into a preset number of equal parts in a top-down manner;

[0040] A mean module is configured to determine a single-channel mean value of a preset number of lowermost lines of each of the equal parts;

[0041] A statistics module is configured to count a number of single-channel mean values greater than a layering threshold value;

[0042] A determination module is configured to determine that a layering interface is recognized when the number is greater than a preset value, and the preset value is less than the preset number of lines.

[0043] The present application also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the resin layer interface recognition method of any of the above when executing the program.

[0044] The present application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program is executable by a processor to implement the resin layer interface recognition method of any of the above.

[0045] The present application also provides a computer program product comprising a computer program, wherein the computer program is executable by a processor to implement the resin layer interface recognition method of any of the above.

[0046] This invention provides a method, apparatus, and device for resin layering interface recognition. The method involves receiving an image recognition command, acquiring image information of the resin to be identified based on the command, selecting a region of interest (ROI) in the image information based on image segmentation, dividing the ROI into preset equal parts from top to bottom, determining the single-channel average of the preset number of rows at the bottom of each part, counting the number of instances where the single-channel average is greater than a layering threshold, and determining that a layering interface has been identified when the number is greater than a preset value (where the preset value is less than the preset number of rows). By using the technical solution of this invention to determine whether a resin is an anion resin through equal parts, the computational load is effectively reduced, thereby effectively solving the problem of lag in the processing terminal and effectively improving the efficiency of resin layering recognition. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0048] Figure 1 This is a schematic flowchart of the resin layer interface identification method provided in an embodiment of the present invention;

[0049] Figure 2 This is a schematic diagram of the structure of the resin layer interface recognition device provided in an embodiment of the present invention;

[0050] Figure 3 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0052] The following is combined with Figures 1-3 This invention describes a method, apparatus, and device for identifying resin layered interfaces.

[0053] Figure 1 This is a schematic flowchart of the resin layer interface identification method provided in an embodiment of the present invention.

[0054] like Figure 1As shown, the resin layered interface recognition method provided by the embodiment of the application is executed by a layered recognition platform, and mainly includes the following steps:

[0055] 101. Receive an image recognition instruction, and collect image information of the resin to be recognized based on the image recognition instruction.

[0056] In a specific implementation process, a Distributed Control System (DCS) issues an image recognition instruction in a sequence control process, at which time the layered recognition platform can receive the image recognition instruction through a Programmable Logic Controller (PLC), and then collect image information of the resin to be recognized based on the image recognition instruction. The image information can be picture information collected directly, or image information obtained from video information collected.

[0057] The layered recognition platform is communicatively connected to the PLC through a network switch and communicates data by using a Snap7 protocol. An I / O port of the PLC is connected to the DCS system by using a hard-wire connection. Thus, the communication between the DCS system and the layered recognition platform is achieved. The PLC, as a hardware isolation between the layered recognition platform and the DCS control system, also effectively reduces the network security risk of the DCS control system.

[0058] 102. Select a region of interest in the image information based on image segmentation.

[0059] After the image information is obtained, it is first segmented, divided into a plurality of mutually non-overlapping regions, and then the pixel features of each region are extracted. When the resin features in the pixel features are recognized, the corresponding region is taken as a region of interest (ROI).

[0060] 103. Divide the region of interest into preset equal parts in a top-to-bottom manner.

[0061] In the process of image recognition, the top-to-bottom manner is defined as the orientation of the front view direction, and thus the region of interest is divided in a top-to-bottom manner, i.e., from the upper part to the lower part of the current interface during image processing, into preset equal parts m, such as 30 equal parts.

[0062] 104. Determine the single-channel mean value of a preset number of lines of the lowest edge of each equal part.

[0063] After the equal division, the single-channel mean value of the preset number n of the lowest rows of each equal part is calculated, wherein the position of the lowest side is described in the same manner from top to bottom as described above. For example, n = 10, the single-channel mean value of the 10 rows of the lowest side of each equal part is calculated.

[0064] 105, the number of single-channel mean values greater than the stratification threshold is counted.

[0065] The number of all single-channel mean values greater than the stratification threshold is counted and recorded as p. The stratification threshold can be adaptively determined according to different resins, so that the final stratification recognition method is more suitable for a variety of different resins, and the recognition accuracy is ensured.

[0066] 106, when the number is greater than the preset value, it is determined that the stratification interface is recognized, and the preset value is less than the preset number of rows.

[0067] When the preset number p is greater than the preset value q, it indicates that the stratification interface is recognized at this time, and the stratification signal can be fed back to the DCS control system at this time to ensure the smooth progress of the subsequent process flow. The preset value q is less than the preset number of rows n. For example, q = 15, indicating that the stratification interface is recognized at this time, and the stratification signal can be fed back at this time. Therefore, by using a similar two-out-of-three voting method, the stratification recognition has a larger fault tolerance space, and the misrecognition rate and the rejection recognition rate can be effectively reduced.

[0068] The resin stratification interface recognition provided in this embodiment receives an image recognition instruction, acquires image information of a resin to be recognized based on the image recognition instruction, selects a region of interest in the image information based on image segmentation, divides the region of interest into a preset number of equal parts in a top-to-bottom manner, determines the single-channel mean value of the preset number of rows of the lowest side of each equal part, counts the number of single-channel mean values greater than the stratification threshold, and determines that the stratification interface is recognized when the number is greater than the preset value, and the preset value is less than the preset number of rows. By using the technical solution of the present application, the calculation amount is effectively reduced by judging whether the equal part is a negative resin, thereby effectively solving the problem of terminal lag and effectively improving the efficiency of resin stratification recognition.

[0069] Further, on the basis of the above-mentioned embodiments, in this embodiment, after selecting the region of interest in the image information based on image segmentation, it further includes: screening the selected target region image of the preset number of frames in the region of interest; obtaining the single-channel average value of the target region image as the positive resin characteristic value; adding a bias to the positive resin characteristic value to obtain a negative resin threshold value, and taking the negative resin threshold value as the layering threshold value. And adding a bias to the positive resin characteristic value to obtain a negative resin threshold value includes: determining the resin properties of different mixed beds and the resin properties of different pollution levels; determining the bias size based on the resin properties of different mixed beds and the resin properties of different pollution levels; adding the bias size to the positive resin characteristic value to obtain the negative resin threshold value.

[0070] Specifically, after receiving the image recognition instruction, the ROI is selected by using the image segmentation technology, and the target region image of the preset number of frames can be screened, for example, the first ten frames of region image can be screened, and then the single-channel average value of the ten frames of target region image is calculated as the positive resin characteristic value. Then, the resin properties of different mixed beds and the resin properties of different pollution levels are determined, and the bias size is determined in an adaptive manner based on the resin properties of different mixed beds and the resin properties of different pollution levels, and the positive resin is subjected to bias processing of the bias size, so as to obtain the negative resin threshold value, which is taken as the layering threshold value. And the layering threshold value determined by this method is not a fixed value, but is obtained by adaptive and flexible adjustment according to different resins. Thus, the accuracy of layering recognition can be ensured for resins with different property characteristics, and the layering recognition of resins of different mixed beds can be applied.

[0071] Further, on the basis of the above-mentioned embodiments, in this embodiment, the image recognition instruction is received, the image information of the resin to be identified is collected based on the image recognition instruction, which includes: receiving the control instruction from the distributed control system through the programmable logic controller; based on the control instruction, the camera collects the image information of the resin to be identified by using the resin tank sight mirror. And when at least one of the following conditions is met: the camera's collection signal is not recognized, the programmable logic controller's control signal is not recognized, the device power failure signal is recognized, and the calculation parameter overrun signal is recognized, the signal transmission between the programmable logic controller and the distributed control system is disconnected; an alarm prompt is issued through the distributed control system.

[0072] Specifically, a high-definition camera is installed at the position of the resin tank sight mirror, and then the DCS system controls the camera to capture images at a speed of 20-30 frames per second, and the collected video is communicated to the layering identification platform through the network switch. The layering identification platform performs image analysis and processing to determine whether there is a layering signal.

[0073] When the hierarchical identification platform fails to detect camera signals, fails to detect control information from the programmable logic controller (PLC), detects excessive calculation parameters, or detects power failures in any of the following faults, it indicates a hardware connection or communication failure. In this case, the normal signals sent by the PLC to the DCS control system will disappear, and the DCS will issue an alarm, effectively reminding the operator to respond promptly. Simultaneously, the hierarchical identification platform automatically records log files with different alarm levels, facilitating maintenance personnel to quickly identify the cause of the fault.

[0074] Furthermore, based on the above embodiments, this embodiment also includes: identifying the signal processing type during the resin layering interface identification process; if the signal processing type is signal reading and threshold calculation, then the main thread is used for data processing; if the signal processing type is function calculation and video display, then the first sub-thread is used for data processing; if the signal processing type is feedback layering signal, then the second sub-thread is used for data processing.

[0075] Specifically, by employing multi-threading technology, the program's response speed can be better guaranteed, effectively shortening image recognition time and thus improving the efficiency of layered recognition. Simplifying image algorithm recognition while reducing data computation ensures the high efficiency and stability of layered recognition.

[0076] Based on the same general inventive concept, this invention also protects a resin layer interface identification device. The resin layer interface identification device provided by this invention will be described below. The resin layer interface identification device described below can be referred to in correspondence with the resin layer interface identification method described above.

[0077] Figure 2 This is a schematic diagram of the structure of the resin layer interface recognition device provided in an embodiment of the present invention.

[0078] like Figure 2 As shown, an embodiment of the present invention provides a resin layer interface identification device, comprising:

[0079] The acquisition module 201 is used to receive image recognition instructions and acquire image information of the resin to be identified based on the image recognition instructions;

[0080] The segmentation module 202 is used to select the region of interest in the image information based on image segmentation;

[0081] The partitioning module 203 is used to divide the region of interest into preset equal parts in a top-to-bottom manner;

[0082] The mean module 204 is used to determine the single-channel mean of the bottom preset row number of each equal part;

[0083] a statistics module 205, configured to count a number of single-channel means greater than a layered threshold value;

[0084] a determination module 206, configured to determine that a layered interface is recognized when the number is greater than a preset value, the preset value being less than the preset number of lines.

[0085] The resin layered interface recognition device provided in the embodiment receives an image recognition instruction, acquires image information of the resin to be recognized based on the image recognition instruction, selects a region of interest in the image information based on image segmentation, divides the region of interest into a preset number of equal parts in a top-down manner, determines a single-channel mean value of a preset number of lines at the bottom of each equal part, counts a number of single-channel means greater than a layered threshold value, and determines that a layered interface is recognized when the number is greater than a preset value, the preset value being less than the preset number of lines. The technical solution of the present application effectively reduces the amount of calculation by judging whether the equal part is a negative resin, thereby effectively solving the problem of terminal lag and effectively improving the efficiency of resin layering recognition.

[0086] Further, the embodiment further includes a threshold value determination module, configured to:

[0087] select a target region image of a preset number of frames in the region of interest;

[0088] obtain a single-channel mean value of the target region image as a positive resin characteristic value;

[0089] add a bias to the positive resin characteristic value to obtain a negative resin threshold value, and use the negative resin threshold value as the layered threshold value.

[0090] Further, the threshold value determination module in the embodiment is further configured to:

[0091] determine resin properties of different mixed beds and resin properties of different pollution levels;

[0092] determine a bias size based on the resin properties of the different mixed beds and the resin properties of the different pollution levels;

[0093] add the bias size to the positive resin characteristic value to obtain a negative resin threshold value.

[0094] Further, the acquisition module 201 in the embodiment is specifically configured to:

[0095] receive a control instruction from a distributed control system through a programmable logic controller;

[0096] based on the control instruction, control the camera to acquire image information of the resin to be recognized using a resin tank viewer.

[0097] Further, the embodiment further includes an alarm module, configured to:

[0098] when at least one of the camera acquisition signal is not identified, the programmable logic controller control signal is not identified, the device power failure signal is identified and the calculation parameter overrun signal is identified, the programmable logic controller is disconnected from the signal transmission of the distributed control system;

[0099] an alarm prompt is sent through the distributed control system.

[0100] Further, the programmable logic controller in the embodiment is connected with the distributed control system through a hard line.

[0101] Further, the embodiment further comprises a thread module, which is used for:

[0102] identifying the signal processing type in the resin layering interface identification process;

[0103] if the signal processing type is signal reading and threshold operation, the main thread is used for data processing;

[0104] if the signal processing type is function operation and video display, the first sub-thread is used for data processing;

[0105] if the signal processing type is feedback layering signal, the second sub-thread is used for data processing.

[0106] Further, the segmentation module 202 in the embodiment is specifically used for:

[0107] dividing the image information into a plurality of mutually non-overlapping regions;

[0108] extracting pixel features of each region;

[0109] when the resin feature in the pixel features is identified, the corresponding region is taken as a region of interest.

[0110] Figure 3 is a structural schematic diagram of an electronic device provided by the embodiment of the present application.

[0111] As Figure 3As shown, the electronic device can include a processor 310, a communications interface 320, a memory 330, and a communications bus 340, wherein the processor 310, the communications interface 320, and the memory 330 complete mutual communication through the communications bus 340. The processor 310 can invoke a logical instruction in the memory 330 to execute a resin layering interface identification method, which includes receiving an image identification instruction, collecting image information of a resin to be identified based on the image identification instruction, selecting a region of interest in the image information based on image segmentation, dividing the region of interest into a preset equal part in a top-to-bottom manner, determining a single-channel mean value of a preset number of lowermost edges of each equal part, counting a number of single-channel mean values greater than a layering threshold value, and determining that a layering interface is identified when the number is greater than a preset value, the preset value being less than the preset number of rows.

[0112] In addition, the logical instruction in the memory 330 described above can be implemented in the form of a software functional unit and sold or used as an independent product, and can be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0113] On the other hand, the present application also provides a computer program product, which includes a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the resin layering interface identification method provided by the above-mentioned methods, which includes: receiving an image identification instruction, collecting image information of a resin to be identified based on the image identification instruction, selecting a region of interest in the image information based on image segmentation, dividing the region of interest into a preset equal part in a top-to-bottom manner, determining a single-channel mean value of a preset number of lowermost edges of each equal part, counting a number of single-channel mean values greater than a layering threshold value, and determining that a layering interface is identified when the number is greater than a preset value, the preset value being less than the preset number of rows.

[0114] In yet another aspect, the present application also provides a non-transitory computer readable storage medium having stored thereon a computer program, which, when executed by a processor, implements the resin layering interface identification method provided by the above method, and the method comprises: receiving an image identification instruction, collecting image information of a resin to be identified based on the image identification instruction; selecting a region of interest in the image information based on image segmentation; dividing the region of interest into a preset equal number of parts in a top-to-bottom manner; determining the single-channel mean value of a preset number of rows at the bottom of each of the equal parts; counting the number of single-channel mean values greater than a layering threshold; and when the number is greater than a preset value, determining that a layering interface is identified, and the preset value is less than the preset number of rows.

[0115] The device embodiments described above are merely illustrative, wherein the units illustrated as separate components can or can not be physically separated, and the components illustrated as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0116] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and necessary universal hardware platforms, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, server, or network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0117] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A resin layer interface recognition method characterized by, The method comprises the following steps: receiving an image recognition instruction, and collecting image information of resin to be recognized based on the image recognition instruction; selecting a region of interest in the image information based on image segmentation; dividing the region of interest into preset equal parts in a top-to-bottom manner; determining the single-channel mean value of a preset number of rows at the bottom of each equal part; counting the number of single-channel mean values greater than a layering threshold value; wherein the setting process of the layering threshold value comprises: screening a preset number of frames of selected target region images in the region of interest; obtaining the single-channel mean value of the target region images as a positive resin characteristic value; adding a bias to the positive resin characteristic value to obtain a negative resin threshold value, and taking the negative resin threshold value as the layering threshold value; when the number is greater than a preset value, it is determined that the layering interface is recognized, and the preset value is less than the preset number of rows; wherein the step of adding a bias to the positive resin characteristic value to obtain a negative resin threshold value comprises: determining the resin properties of different mixed beds and the resin properties of different pollution levels; determining the bias size based on the resin properties of different mixed beds and the resin properties of different pollution levels; adding the bias size to the positive resin characteristic value to obtain a negative resin threshold value.

2. The resin layering interface identification method of claim 1, wherein, The step of receiving an image recognition instruction and collecting image information of resin to be recognized based on the image recognition instruction comprises: receiving a control instruction from a distributed control system through a programmable logic controller; controlling the camera to collect image information of resin to be recognized using a resin tank viewer based on the control instruction.

3. The resin layering interface identification method of claim 2, wherein, Further comprising: when at least one of the following conditions is met: no signal collected by the camera is recognized, no control signal of the programmable logic controller is recognized, a device power failure signal is recognized, and a calculation parameter overrun signal is recognized, disconnecting the signal transmission between the programmable logic controller and the distributed control system; issuing an alarm prompt through the distributed control system.

4. The resin layering interface identification method of claim 2, wherein, The programmable logic controller and the distributed control system are connected through a hardwire.

5. The resin layer interface identification method according to any one of claims 1 to 4, characterized by, Further comprising: identifying the signal processing type in the resin layering interface recognition process; if the signal processing type is signal reading and threshold operation, a main thread is used for data processing; if the signal processing type is function operation and video display, a first sub-thread is used for data processing; if the signal processing type is a feedback layering signal, a second sub-thread is used for data processing.

6. The resin layer interface identification method according to any one of claims 1 to 4, characterized by, The step of selecting a region of interest in the image information based on image segmentation comprises: dividing the image information into a plurality of mutually non-overlapping regions; extracting pixel features of each region; when a resin feature in the pixel features is recognized, the corresponding region is taken as the region of interest.

7. A resin layer interface recognition apparatus characterized by comprising: The method comprises the following steps: a collection module for receiving an image recognition instruction and collecting image information of resin to be recognized based on the image recognition instruction; a segmentation module for selecting a region of interest in the image information based on image segmentation; a division module for dividing the region of interest into preset equal parts in a top-to-bottom manner; a mean value module for determining the single-channel mean value of a preset number of rows at the bottom of each equal part; The statistical module is configured to count the number of single-channel mean values greater than a layered threshold value; wherein the setting process of the layered threshold value comprises: screening a pre-set number of selected target region images in the region of interest; calculating the single-channel mean value of the target region images as a positive resin characteristic value; determining resin properties of different mixed beds and resin properties of different pollution levels; determining a bias size based on the resin properties of different mixed beds and the resin properties of different pollution levels; and adding the bias size to the positive resin characteristic value to obtain a negative resin threshold value; The determination module is configured to determine that a layered interface is recognized when the number is greater than a pre-set value, and the pre-set value is less than the pre-set number of rows.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor implements the resin layered interface recognition method according to any one of claims 1 to 6 when executing the program.

Images