Cloth window deviation early warning method and system and electronic equipment

Abstract

The application provides a cloth window offset early warning method, system and electronic equipment. The method comprises the steps of: judging whether a first offset early warning condition is established, the condition being that in the current cloth cutting operation, the current cloth window passing frequency is within the cloth window passing frequency range and the vacuum adsorption device running time is within the running time range; if the first offset early warning condition is established, judging whether a second offset early warning condition is established, the condition being that the to-be-identified window passing cloth image has a pre-offset feature, and the current adsorption force of the vacuum adsorption device reaches a predetermined adsorption force threshold, the to-be-identified window passing cloth image being a continuous window passing cloth image within a predetermined time period obtained in response to the judgment result of the establishment of the first offset early warning condition; and if the second offset early warning condition is established, outputting a cloth window offset early warning signal. The method comprises functional modules corresponding to the implementation of the above steps. The electronic equipment: the processor implements the method when executing the computer program saved in the memory. According to the application, the problem that the subsequent cloth cutting quality is affected due to the cloth window offset during the automatic cutting bed operation can be solved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of cloth passing through the window monitoring, and more particularly, relates to a cloth passing through the window offset early warning method, system and electronic equipment applied to an automatic cutting bed. BACKGROUND

[0002] The automatic cutting bed is an automatic device for cutting cloth to obtain a cutting piece, mainly including a cutting table, a cutting head, a picking table and a cloth passing through the window mechanism. The cutting head is used for cutting the cloth on the cutting table. The cloth passing through the window mechanism is used for transferring the cut cloth on the cutting table to the picking table for the next picking and material distribution. Specifically, the cloth passing through the window mechanism includes a conveying device and a vacuum adsorption device. The conveying device is used to realize the transmission of the cloth between the cutting table and the picking table, and the vacuum adsorption device is used to provide adsorption force to the cloth during the cloth passing through the window to make it adsorb on the conveying device.

[0003] However, in the actual cloth cutting operation, the cloth passing through the window may be offset intermittently. When the cloth cut by the automatic cutting bed is continuous, the offset cloth passing through the window will cause the cloth waiting to be cut on the cutting table to be offset accordingly, thereby affecting the cutting quality of this part of cloth. Therefore, it is necessary to propose a cloth passing through the window offset early warning method to make a corresponding early warning before the cloth passing through the window is offset, so as to avoid the cutting quality of the subsequent cloth being affected. SUMMARY

[0004] The purpose of the present application is to solve the problem that the cutting quality of the subsequent cloth is affected due to the cloth passing through the window offset during the operation of the automatic cutting bed.

[0005] In order to achieve the above purpose, the present application provides a cloth passing through the window offset early warning method, system and electronic equipment applied to an automatic cutting bed.

[0006] According to the first aspect of the present application, a cloth passing through the window offset early warning method applied to an automatic cutting bed is provided, wherein the automatic cutting bed includes a conveying device and a vacuum adsorption device. The conveying device is used to realize the cloth passing through the window. The vacuum adsorption device is used to provide adsorption force to the cloth during the cloth passing through the window to make it adsorb on the conveying device.

[0007] The cloth passing through the window offset early warning method includes:

[0008] determining whether a first offset early warning condition is established, the first offset early warning condition being that in the current cloth cutting operation, the current cloth passing through the window times are within a predetermined cloth passing through the window times range and the running time of the vacuum adsorption device is within a predetermined running time range;

[0009] in response to a result of the determination that the first offset warning condition is established, determining whether a second offset warning condition is established, the second offset warning condition being:

[0010] the image of the fabric passing through the window to be identified is identified as having the pre-offset feature by the pre-constructed pre-offset feature identification model, and the current suction force of the vacuum suction device reaches a predetermined suction force threshold,

[0011] the image of the fabric passing through the window to be identified is a continuous image of the fabric passing through the window within a predetermined time period obtained in response to a result of the determination that the first offset warning condition is established;

[0012] in response to a result of the determination that the second offset warning condition is established, outputting a fabric passing through the window offset warning signal.

[0013] Optionally, before determining whether the first offset warning condition is established, the method further comprises:

[0014] determining, according to the property of the fabric and based on a pre-obtained mapping relationship between the fabric property, the fabric passing through the window frequency range and the vacuum suction device running time range, the fabric passing through the window frequency range and the vacuum suction device running time range corresponding to the fabric.

[0015] Optionally, the method of obtaining the mapping relationship between the fabric property, the fabric passing through the window frequency range and the vacuum suction device running time range comprises:

[0016] obtaining offset history data of fabrics of different properties, the offset history data including the fabric passing through the window frequency and the vacuum suction device running time when the offset occurs multiple times;

[0017] determining, according to the offset history data, the fabric passing through the window frequency range and the vacuum suction device running time range when the offset occurs for fabrics of different properties.

[0018] Optionally, the method of determining, according to the offset history data, the fabric passing through the window frequency range and the vacuum suction device running time range when the offset occurs for fabrics of different properties comprises:

[0019] for each property of the fabric, obtaining a probability distribution of the fabric passing through the window frequency when the offset occurs multiple times, and determining, according to the probability distribution of the fabric passing through the window frequency, a fabric passing through the window frequency range corresponding to a predetermined first probability interval;

[0020] for each property of the fabric, obtaining a probability distribution of the vacuum suction device running time when the offset occurs multiple times, and determining, according to the probability distribution of the vacuum suction device running time, a vacuum suction device running time range corresponding to a predetermined second probability interval.

[0021] Optionally, the method for constructing the pre-offset feature recognition model includes:

[0022] Acquire historical data of fabrics with different properties before the first offset. The historical data before the first offset includes continuous windowed fabric images within a predetermined time period before multiple windowed offsets occur.

[0023] The fabric attribute information and consecutive windowed fabric images within a predetermined time period before each windowed shift are used as samples. The samples with pre-shift features are used as labels to train the neural network model to obtain the pre-shift feature recognition model.

[0024] Optionally, before determining whether the second offset warning condition is met, the method further includes:

[0025] The adsorption threshold corresponding to the fabric is determined based on the properties of the fabric and the pre-acquired mapping relationship between the fabric properties and the adsorption threshold.

[0026] Optionally, the method for obtaining the mapping relationship between the fabric properties and the adhesion threshold includes:

[0027] Acquire historical data of fabrics with different properties before the second offset, including the adsorption force change curves within a predetermined time before multiple window offsets.

[0028] For each type of fabric, the real-time value of the adsorption force corresponding to the midpoint of the predetermined time period is determined based on the adsorption force change curve within the predetermined time period before each window offset, and the average value of the multiple real-time adsorption force values ​​is used as the adsorption force threshold corresponding to that type of fabric.

[0029] Optionally, the fabric properties include the fabric material and thickness.

[0030] According to a second aspect of the present invention, a fabric window offset early warning system for use in an automatic cutting bed is provided, the automatic cutting bed including a conveying device and a vacuum adsorption device, the conveying device being used to realize fabric window passing, and the vacuum adsorption device being used to provide adsorption force to the fabric during the fabric window passing process so that it is adsorbed on the conveying device.

[0031] The fabric shift warning system includes the following functional modules:

[0032] The first judgment module is used to determine whether the first offset warning condition is met. The first offset warning condition is that in this fabric cutting operation, the current number of fabric window passes is within the predetermined number of fabric window passes and the running time of the vacuum adsorption device is within the predetermined running time range.

[0033] The second judgment module is used to determine whether the second offset warning condition is met in response to the judgment result that the first offset warning condition is met. The second offset warning condition is:

[0034] The fabric image to be identified is recognized as having pre-offset features by a pre-constructed pre-offset feature recognition model, and the current adsorption force of the vacuum adsorption device reaches a predetermined adsorption force threshold.

[0035] The windowed fabric images to be identified are continuous windowed fabric images within a predetermined time period obtained in response to the judgment result that the first offset warning condition is met.

[0036] The offset warning module is used to output a fabric offset warning signal in response to the judgment result that the second offset warning condition is met.

[0037] According to a third aspect of the present invention, an electronic device is provided, comprising a processor and a memory, wherein the processor executes a computer program stored in the memory to implement any of the above-described fabric window offset warning methods applied to an automatic cutting bed.

[0038] The beneficial effects of this invention are as follows:

[0039] The present invention provides a fabric window offset early warning method for automatic cutting beds. First, it determines whether a first offset early warning condition is met. The first offset early warning condition is that, in the current fabric cutting operation, the current number of fabric window passes is within a predetermined range, and the running time of the vacuum adsorption device is within a predetermined running time range. Second, in response to the determination that the first offset early warning condition is met, it determines whether a second offset early warning condition is met. The second offset early warning condition is that the fabric image to be identified is recognized as having pre-offset features by a pre-constructed pre-offset feature recognition model, and the current adsorption force of the vacuum adsorption device reaches a predetermined adsorption force threshold. The fabric image to be identified is a series of consecutive fabric images within a predetermined time period obtained in response to the determination that the first offset early warning condition is met. Finally, in response to the determination that the second offset early warning condition is met, a fabric window offset early warning signal is output.

[0040] The fabric window shift early warning method of the present invention, applied to an automatic cutting bed, acquires relevant data of a first shift early warning condition in real time during the automatic cutting bed operation, namely the current number of fabric window shifts and the running time of the vacuum adsorption device. It monitors this data and, if the current number of fabric window shifts is within a predetermined range and the running time of the vacuum adsorption device is within a predetermined running time range, then a step is initiated to determine whether a second shift early warning condition is met. Specifically, it determines whether consecutive fabric images within a predetermined time period acquired in response to the determination that the first shift early warning condition is met retain the characteristics before the shift and whether the current adsorption force of the vacuum adsorption device reaches a predetermined adsorption force threshold. If the second shift early warning condition is met, an early warning mechanism is activated. Therefore, the fabric window shift early warning method of the present invention can effectively predict and warn of fabric window shifts, thereby effectively solving the problem of subsequent fabric cutting quality being affected by fabric window shifts during automatic cutting bed operations.

[0041] The fabric window offset warning system and electronic device of the present invention for automatic cutting beds belong to the same general inventive concept as the fabric window offset warning method for automatic cutting beds described above, and have at least the same beneficial effects as the fabric window offset warning method for automatic cutting beds described above, the beneficial effects of which will not be repeated here.

[0042] Other features and advantages of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0043] The present invention can be better understood by referring to the following description taken in conjunction with the accompanying drawings, in which the same or similar reference numerals are used throughout the drawings to denote the same or similar parts.

[0044] Figure 1 A flowchart illustrating the implementation of a fabric window offset early warning method for an automatic cutting bed according to an embodiment of the present invention is shown.

[0045] Figure 2 A structural block diagram of a fabric window offset early warning system applied to an automatic cutting bed according to an embodiment of the present invention is shown. Detailed Implementation

[0046] To enable those skilled in the art to more fully understand the technical solutions of the present invention, exemplary embodiments of the present invention will be described more comprehensively and in detail below with reference to the accompanying drawings. Obviously, the one or more embodiments of the present invention described below are merely one or more specific ways to implement the technical solutions of the present invention, and are not exhaustive. It should be understood that other ways belonging to a general inventive concept can be used to implement the technical solutions of the present invention, and should not be limited to the embodiments described exemplary. Based on one or more embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0047] Example: Figure 1 A flowchart illustrating the implementation of a fabric window offset early warning method for an automatic cutting bed according to an embodiment of the present invention is shown. (Refer to...) Figure 1 The fabric window offset early warning method applied to an automatic cutting bed according to an embodiment of the present invention includes the following steps:

[0048] Step S100: Determine whether the first offset warning condition is met. The first offset warning condition is that in this fabric cutting operation, the current number of fabric window passes is within the predetermined number of fabric window passes and the running time of the vacuum adsorption device is within the predetermined running time range.

[0049] Step S200: In response to the judgment result that the first offset warning condition is met, determine whether the second offset warning condition is met. The second offset warning condition is:

[0050] The fabric image to be identified through the window is recognized as having pre-offset features by a pre-built pre-offset feature recognition model, and the current adsorption force of the vacuum adsorption device reaches a predetermined adsorption force threshold.

[0051] The windowed fabric images to be identified are continuous windowed fabric images within a predetermined time period obtained in response to the judgment result that the first offset warning condition is met.

[0052] Step S300: In response to the judgment result that the second offset warning condition is met, output the fabric window offset warning signal.

[0053] Specifically, in this embodiment of the invention, "fabric passing through the window" refers to the process of transferring the cut fabric from the cutting table to the picking table.

[0054] Specifically, in this embodiment of the invention, the recognition principle of the pre-offset feature recognition model is as follows:

[0055] Each windowed fabric image in a series of windowed fabric images within a predetermined time period is preprocessed to obtain a direction line representing the fabric transport direction corresponding to each windowed fabric image.

[0056] The obtained direction lines are fitted based on the time information corresponding to each window fabric image to obtain the fitted direction lines that characterize the transmission direction of the window fabric within a predetermined time period.

[0057] Feature recognition is performed on the obtained fitted direction lines to determine whether they have the corresponding pre-offset features.

[0058] Specifically, the fabric window shift warning method applied to an automatic cutting bed according to this embodiment of the invention first determines whether the current number of fabric window shifts is within a predetermined range and whether the running time of the vacuum adsorption device is within a predetermined running time range. If so, it then determines whether the fabric image to be identified has pre-shift features as recognized by the pre-constructed pre-shift feature recognition model, and whether the current adsorption force of the vacuum adsorption device reaches a predetermined adsorption force threshold. The reason for this setting is that the frequency of fabric window shifts is low, that is, the time interval between two fabric window shift events is relatively long. If a step to determine whether the first shift warning condition is met is not set, it is necessary to collect fabric window images in real time and perform corresponding image processing and recognition. The corresponding image acquisition action will generate a large amount of image data stream, resulting in a considerable amount of image data processing, which is obviously uneconomical and places very high demands on the relevant hardware. Therefore, in this embodiment of the invention, a step is set to determine whether the first offset warning condition is met, that is, a trigger condition for image acquisition and recognition is set. After determining that the first offset warning condition is met, it is only necessary to acquire continuous images of the fabric through the window within a predetermined time period and perform corresponding image processing and recognition, thereby greatly reducing the image data stream and related image data processing workload.

[0059] Furthermore, step S100 of this embodiment of the invention, before determining whether the first offset warning condition is met, further includes:

[0060] Based on the properties of the fabric, and based on the mapping relationship between the pre-acquired fabric properties, the range of fabric windowing times, and the range of vacuum adsorption device operating time, the range of fabric windowing times and the range of vacuum adsorption device operating time are determined.

[0061] Furthermore, in this embodiment of the invention, the method for obtaining the mapping relationship between fabric properties, the range of fabric window passes, and the range of vacuum adsorption device operating time includes:

[0062] Obtain historical offset data for fabrics with different properties. The historical offset data includes the number of times the fabric crossed the window when multiple window offsets occurred and the running time of the vacuum adsorption device.

[0063] Based on historical offset data, determine the range of fabric window crossing times and the range of vacuum adsorption device operating time when fabrics of different properties experience window crossing offset.

[0064] Furthermore, in this embodiment of the invention, determining the range of fabric window crossing times and the range of vacuum adsorption device operating time when different attribute fabrics experience window crossing offset based on historical offset data includes:

[0065] For each type of cloth, obtain the probability distribution of the number of times the cloth crosses the window when it has multiple window offsets, and determine the range of the number of times the cloth crosses the window corresponding to the predetermined first probability interval based on the probability distribution of the number of times the cloth crosses the window.

[0066] For each type of fabric, obtain the probability distribution of the vacuum adsorption device runtime when it experiences multiple window offsets, and determine the range of vacuum adsorption device runtime corresponding to the predetermined second probability interval based on the probability distribution of the vacuum adsorption device runtime.

[0067] Specifically, in this embodiment of the invention, by analyzing the historical offset data of fabrics with different properties, the range of fabric window crossing times and the range of vacuum adsorption device runtime when fabrics with different properties experience window crossing offset are obtained. The range of fabric window crossing times when fabric window crossing offset occurs is not a range covering all fabric window crossing times, but rather a range of fabric window crossing times that are prone to occur, determined based on the probability distribution of fabric window crossing times. Similarly, the range of vacuum adsorption device runtime when fabric window crossing offset occurs is a range of vacuum adsorption device runtimes that are prone to occur, and does not cover all vacuum adsorption device runtimes.

[0068] Furthermore, in step S200 of this embodiment of the invention, the method for constructing the pre-offset feature recognition model includes:

[0069] Acquire historical data of fabrics with different properties before the first offset. The historical data before the first offset includes continuous windowed fabric images within a predetermined time period before multiple windowed offsets occur.

[0070] The fabric attribute information and consecutive windowed fabric images within a predetermined time period before each windowed shift are used as samples. The samples with pre-shift features are used as labels to train the neural network model to obtain the pre-shift feature recognition model.

[0071] Furthermore, step S200 of this embodiment of the invention, before determining whether the second offset warning condition is met, further includes:

[0072] The adsorption threshold of the fabric is determined based on the fabric properties and the mapping relationship between the pre-acquired fabric properties and the adsorption threshold.

[0073] Furthermore, in this embodiment of the invention, the method for obtaining the mapping relationship between fabric properties and the adsorption force threshold includes:

[0074] Acquire historical data of fabrics with different properties before the second offset. The historical data before the second offset includes the adsorption force change curves within a predetermined time before multiple window offsets.

[0075] For each type of fabric, the real-time value of the adsorption force corresponding to the midpoint of the predetermined time period is determined based on the adsorption force change curve within the predetermined time period before each window offset, and the average value of the multiple real-time adsorption force values ​​is used as the adsorption force threshold corresponding to that type of fabric.

[0076] Furthermore, in this embodiment of the invention, the properties of the fabric include the material and thickness of the fabric.

[0077] Accordingly, based on the fabric window offset early warning method applied to automatic cutting beds in the embodiments of the present invention, the embodiments of the present invention also propose a fabric window offset early warning system applied to automatic cutting beds.

[0078] Figure 2 A structural block diagram of a fabric window offset early warning system applied to an automatic cutting bed according to an embodiment of the present invention is shown. (Refer to...) Figure 2 The fabric window offset early warning system for automatic cutting beds according to embodiments of the present invention includes the following functional modules:

[0079] The first judgment module is used to determine whether the first offset warning condition is met. The first offset warning condition is that in this fabric cutting operation, the current number of fabric window passes is within the predetermined number of fabric window passes and the running time of the vacuum adsorption device is within the predetermined running time range.

[0080] The second judgment module is used to determine whether the second offset warning condition is met in response to the judgment result that the first offset warning condition is met. The second offset warning condition is:

[0081] The fabric image to be identified through the window is recognized as having pre-offset features by a pre-built pre-offset feature recognition model, and the current adsorption force of the vacuum adsorption device reaches a predetermined adsorption force threshold.

[0082] The windowed fabric images to be identified are continuous windowed fabric images within a predetermined time period obtained in response to the judgment result that the first offset warning condition is met.

[0083] The offset warning module is used to output a fabric offset warning signal in response to the judgment result that the second offset warning condition is met.

[0084] Accordingly, based on the fabric window offset early warning method for automatic cutting beds in the embodiments of the present invention, the embodiments of the present invention also propose an electronic device, which includes a processor and a memory. When the processor executes the computer program stored in the memory, it implements the fabric window offset early warning method for automatic cutting beds in the embodiments of the present invention.

[0085] While one or more embodiments of the present invention have been described above, those skilled in the art will recognize that the present invention can be implemented in any other form without departing from its spirit and scope. Therefore, the embodiments described above are illustrative and not restrictive, and many modifications and substitutions will be apparent to those skilled in the art without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A fabric window offset early warning method applied to an automatic cutting bed, the automatic cutting bed including a conveying device and a vacuum adsorption device, the conveying device being used to realize fabric window passing, and the vacuum adsorption device being used to provide adsorption force to the fabric during the fabric window passing process so that it is adsorbed on the conveying device. Its features are, The fabric shift warning method includes: Determine whether the first offset warning condition is met. The first offset warning condition is that in this fabric cutting operation, the current number of fabric window passes is within the predetermined number of fabric window passes and the running time of the vacuum adsorption device is within the predetermined running time range. In response to the determination result that the first offset warning condition is met, it is determined whether the second offset warning condition is met. The second offset warning condition is: The fabric image to be identified is recognized as having pre-offset features by a pre-constructed pre-offset feature recognition model, and the current adsorption force of the vacuum adsorption device reaches a predetermined adsorption force threshold. The windowed fabric images to be identified are continuous windowed fabric images within a predetermined time period obtained in response to the judgment result that the first offset warning condition is met. In response to the judgment result that the second offset warning condition is met, a fabric window offset warning signal is output; The method for constructing the pre-offset feature recognition model includes: Acquire historical data of fabrics with different properties before the first offset. The historical data before the first offset includes continuous windowed fabric images within a predetermined time period before multiple windowed offsets occur. The fabric attribute information and the continuous windowed fabric images within a predetermined time period before each windowed shift of the corresponding fabric are used as samples. The samples with pre-shift features are used as labels to train the neural network model to obtain the pre-shift feature recognition model. Before determining whether the second offset warning condition is met, the method further includes: The adsorption force threshold corresponding to the fabric is determined based on the properties of the fabric and the pre-acquired mapping relationship between the fabric properties and the adsorption force threshold. The method for obtaining the mapping relationship between the fabric properties and the adsorption force threshold includes: Acquire historical data of fabrics with different properties before the second offset, including the adsorption force change curves within a predetermined time before multiple window offsets. For each type of fabric, the real-time value of the adsorption force corresponding to the midpoint of the predetermined time period is determined based on the adsorption force change curve within the predetermined time period before each window shift, and the average value of the multiple real-time adsorption force values ​​is used as the adsorption force threshold corresponding to the fabric of that type of attribute. The specific recognition principle of the pre-offset feature recognition model is as follows: Each windowed fabric image in a series of windowed fabric images within a predetermined time period is preprocessed to obtain a direction line representing the fabric transport direction corresponding to each windowed fabric image. The obtained direction lines are fitted based on the time information corresponding to each window fabric image to obtain the fitted direction lines that characterize the transmission direction of the window fabric within a predetermined time period. Feature recognition is performed on the obtained fitted direction lines to determine whether they have the corresponding pre-offset features.

2. The fabric window offset early warning method applied to an automatic cutting bed according to claim 1, characterized in that, Before determining whether the first offset warning condition is met, the method further includes: Based on the properties of the fabric, and based on the mapping relationship between the pre-acquired fabric properties, the range of fabric window passing times, and the range of vacuum adsorption device operating time, the range of fabric window passing times and the range of vacuum adsorption device operating time are determined.

3. The fabric window offset early warning method applied to an automatic cutting bed according to claim 2, characterized in that, The method for obtaining the mapping relationship between the fabric properties, the range of fabric window passes, and the operating time range of the vacuum adsorption device includes: Obtain historical offset data of fabrics with different properties, including the number of times the fabric passed through the window and the running time of the vacuum adsorption device when multiple window offsets occurred. Based on the historical offset data, determine the range of fabric window crossing times and the range of vacuum adsorption device operating time when fabrics with different properties experience window crossing offset.

4. The fabric window offset early warning method applied to an automatic cutting bed according to claim 3, characterized in that, The determination of the range of fabric window crossing times and the range of vacuum adsorption device operating time when different attribute fabrics experience window crossing offset based on the offset history data includes: For each type of fabric, obtain the probability distribution of the number of times the fabric passes through the window when it has multiple window offsets, and determine the range of the number of times the fabric passes through the window corresponding to the predetermined first probability interval based on the probability distribution of the number of times the fabric passes through the window. For each type of fabric, obtain the probability distribution of the vacuum adsorption device runtime when it experiences multiple window offsets, and determine the vacuum adsorption device runtime range corresponding to the predetermined second probability interval based on the probability distribution of the vacuum adsorption device runtime.

5. The fabric window offset early warning method applied to an automatic cutting bed according to claim 4, characterized in that, The properties of fabric include its material and thickness.

6. A fabric offset warning system for use in automatic cutting beds, characterized in that, Used to implement the fabric window offset early warning method for automatic cutting beds as described in any one of claims 1-5; The automatic cutting bed includes a conveying device and a vacuum adsorption device. The conveying device is used to realize the fabric passing through the window, and the vacuum adsorption device is used to provide adsorption force to the fabric during the fabric passing through the window so that it is adsorbed on the conveying device. The fabric shift warning system includes: The first judgment module is used to determine whether the first offset warning condition is met. The first offset warning condition is that in this fabric cutting operation, the current number of fabric window passes is within the predetermined number of fabric window passes and the running time of the vacuum adsorption device is within the predetermined running time range. The second judgment module is used to determine whether the second offset warning condition is met in response to the judgment result that the first offset warning condition is met. The second offset warning condition is: The fabric image to be identified is recognized as having pre-offset features by a pre-constructed pre-offset feature recognition model, and the current adsorption force of the vacuum adsorption device reaches a predetermined adsorption force threshold. The windowed fabric images to be identified are continuous windowed fabric images within a predetermined time period obtained in response to the judgment result that the first offset warning condition is met. The offset warning module is used to output a fabric offset warning signal in response to the judgment result that the second offset warning condition is met.

7. An electronic device, characterized in that, The device includes a processor and a memory, wherein the processor executes a computer program stored in the memory to implement the fabric window offset early warning method for an automatic cutting bed as described in any one of claims 1-5.

Images