Detection method for spinning workshop, apparatus, electronic device, and storage medium

Abstract

To provide a detection method for spinning workshop, an apparatus, an electronic device and a storage medium.SOLUTION: A method includes: performing image collection on a process control device of a spinning box in a spinning workshop to obtain an image to be processed; extracting a first image feature from the image to be processed; and processing the first image feature based on a decoder network to obtain a fault detection result. The decoder network comprises a plurality of decoder modules connected in series in sequence. Each decoder module includes a decoder layer and an adaptive classification head. The adaptive classification head is configured to perform classified prediction on an output feature of the decoder layer to obtain a first fault classification result. The fault detection result output by the decoder network includes a second fault classification result output by a last-layer decoder module of the decoder network. The method enables automatic monitoring in the spinning workshop.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to the field of data processing, and in particular to technical fields such as artificial intelligence and computer vision.

Background Art

[0002] In the industrial scenario of the spinning process, the spinning box in the spinning workshop plays a very important role in the manufacturing process of chemical fiber products. However, due to the low efficiency of manual inspection of the spinning box, how to efficiently conduct a patrol inspection of the spinning workshop is a problem currently faced.

Summary of the Invention

Problems to be Solved by the Invention

[0003] The present disclosure provides a detection method, device, electronic device, and storage medium for a spinning workshop to solve or alleviate one or more technical problems in the prior art.

Means for Solving the Problems

[0004] In a first aspect, the present disclosure provides a detection method for a spinning workshop, the method including: performing image acquisition on the process control device of the spinning box in the spinning workshop to obtain an image of the object to be processed; extracting first image features from the image of the object to be processed; processing the first image features based on a decoder network to obtain a fault detection result for the process control device, wherein the decoder network includes a plurality of decoder modules connected in series in sequence, for each decoder module, the decoder module includes a decoder layer and an adaptive classification head, and the adaptive classification head is used to classify and predict the output features of the decoder layer to obtain a first fault classification result, and update the first fault classification result based on a preset classification calibration matrix to obtain a second fault classification result output by the decoder module. The fault detection result output by the decoder network includes the second fault classification result output by the decoder module of the last layer of the decoder network and the fault location output by the decoder module of the last layer.

[0005] In a second aspect, the present disclosure provides a detection device for a spinning workshop, and the device includes: a collection unit configured to collect images of a process control device of a spinning box in a spinning workshop to obtain an image to be processed; an extraction unit configured to extract first image features from the image to be processed; a processing unit configured to process the first image features based on a decoder network to obtain a fault detection result for the process control device. The decoder network includes a plurality of decoder modules connected in series in sequence. For each decoder module, the decoder module includes a decoder layer and an adaptive classification head. The adaptive classification head is used to classify and predict the output features of the decoder layer to obtain a first fault classification result, and update the first fault classification result based on a preset classification calibration matrix to obtain a second fault classification result output by the decoder module. The fault detection result output by the decoder network includes the second fault classification result output by the decoder module of the last layer of the decoder network and the fault location output by the decoder module of the last layer.

[0006] In a third aspect, the present disclosure provides an electronic device, and the device includes: at least one processor; a memory communicatively connected to the at least one processor. Instructions executable by the at least one processor are stored in the memory, and when executed by the at least one processor, cause any one of the methods in the embodiments of the present disclosure to be executed.

[0007] ​​In a fourth aspect, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute any one of the methods in the embodiments of the present disclosure.

[0008] In a fifth aspect, there is provided a program product including a program for executing any one of the methods in the embodiments of the present disclosure when executed by a processor.

[0009] In the embodiments of the present disclosure, it is realized by automating the monitoring of spinning boxes in a spinning workshop, which helps to timely discover the failure areas and failure types.

[0010] It should be understood that the content described herein is not intended to describe the key points or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. For other features of the present disclosure, understanding is promoted through the following description.

Brief Description of the Drawings

[0011] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the plurality of accompanying drawings indicate the same or similar components or elements. These accompanying drawings are not necessarily drawn to scale. It should be understood that these drawings show only some embodiments provided by the present disclosure and should not be regarded as limiting the scope of the present disclosure.

[0012]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Embodiments for Carrying Out the Invention

[0013] Hereinafter, the present disclosure will be described in more detail with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent functionally identical or similar elements. Also, in the accompanying drawings, various aspects of the embodiments are shown, but these accompanying drawings are not necessarily drawn to scale unless otherwise stated.

[0014] Furthermore, for a better explanation of the present disclosure, many specific details are described in the following specific embodiments. Those skilled in the art should understand that the present disclosure can be implemented similarly even without specific details. In some embodiments, well-known methods, means, components, and circuits, etc. that are well-known to those skilled in the art are not described in detail so as not to obscure the gist of the present disclosure.

[0015] Also, the terms "first" and "second" are used only for descriptive purposes and are not to be understood as indicating or implying relative importance, nor are they to implicitly specify the number of the indicated technical features. Therefore, the features defined by the terms "first" and "second" may explicitly or implicitly include one or more of the said features. In the description of the present disclosure, "one or more" means two or more unless otherwise explicitly and specifically limited.

[0016] In the industrial scenario of the spinning process, the spinning box in the spinning workshop plays a very important role in the manufacturing process of chemical fiber products. The important components of the spinning box include the spinning component and the process control device on the back of the spinning component. The important parts of the process control device in the embodiments of the present disclosure include a heating device, a melt distribution pipe, an auxiliary material transport pipe, a metering pump, etc. Here, the melt distribution pipe has a heat preservation layer, and the heating device needs to ensure the temperature of the melt in the melt distribution pipe in order to facilitate the distribution of the melt to the metering pump. The metering pump evenly diverts the melt to each component of the spinning box, forms a thin stream of the melt through the pores of the spinneret in the spinning component, and after the thin stream of the melt exits the spinneret, it is formed into a thread by natural cooling.

[0017] Here, the melt needs to be heat-insulated in the pipes of the spinning box to ensure a temperature that is as uniform as possible.

[0018] The embodiments of the present disclosure mainly relate to the fault detection of the process control device on the back of the spinning component, especially the detection of the heating device, the melt distribution pipe, and the auxiliary material transport pipe. For the sake of easy explanation and understanding, in the embodiments of the present disclosure, the heating device and the melt distribution pipe are collectively referred to as the temperature control system.

[0019] In addition, the embodiments of the present disclosure can also realize the detection of the open area on the back of the spinning component, etc., in order to improve the detection efficiency of the spinning workshop.

[0020] In addition, the main types of the spun products according to the solutions of the embodiments of the present disclosure, and the main types of the filaments according to the embodiments of the present disclosure, may include one or more of Partially Oriented Yarns (POY), Fully Drawn Yarns (FDY), Polyester staple fiber, etc. For example, specifically as the types of yarns, it may include Polyester Partially Oriented Yarns, Polyester Fully Drawn Yarns, Polyester Drawn Yarns, Polyester staple fiber, etc.

[0021] In order to automatically and accurately detect the spinning workshop, in the embodiments of the present disclosure, DETR (DEtection Transformer, a target detection architecture) is used to propose a detection method for the spinning workshop as shown in FIG. 1, including the following.

[0022] In S101, image collection is performed on the process control device of the spinning box in the spinning workshop to obtain the image to be processed.

[0023] Here, a drone can be used to perform image collection on the process control device of the spinning box in the spinning workshop, and the process control device can include a temperature control system and an auxiliary material transport pipe. The temperature control system includes a melt distribution pipe, and its outer layer is covered with a heat preservation layer. The melt distribution pipe is used to transport the melt generated by the final polycondensation reactor from the pipe to the metering pump, and then evenly distribute it to the spinning components of the spinning box for processing. Since the melt distribution pipe requires heat preservation treatment, during detection, the image of the melt distribution pipe can be collected to determine the presence or absence of damage to the heat preservation layer, and further ensure the temperature of the melt distribution pipe. Among them, the auxiliary material transport pipe is used for transporting industrial auxiliary materials.

[0024] The temperature control system further includes a heating device. The heating device needs to keep the safety gate closed during operation.

[0025] In S102, the first image feature is extracted from the image to be processed.

[0026] In S103, based on the decoder network, the first image feature is processed to obtain a fault detection result for the process control device.

[0027] Here, the decoder network includes a plurality of decoder modules connected in series in sequence. For each decoder module, the decoder module includes a decoder layer and an adaptive classification head, and the adaptive classification head is used to classify and predict the output features of the decoder layer to obtain a first fault classification result, and update the first fault classification result based on a preset classification calibration matrix to obtain a second fault classification result output by the decoder module. The fault detection result output by the decoder network includes the second fault classification result output by the decoder module of the last layer of the decoder network and the fault position output by the decoder module of the last layer.

[0028] Here, as shown in FIG. 2, the decoder network shows n decoder modules, where n is a positive integer. In FIG. 2, decoder module 1, decoder module 2, …, decoder module n are connected in series in sequence. Decoder module 1 includes a decoder layer 1 and an adaptive classification head 1. Each decoder layer can include a self-attention module and a cross-attention module (not shown) connected after the self-attention module. Each decoder module after the first decoder module in the decoder network constructs the necessary query vector (Q), key vector (K), and value vector (V) in the decoder module based on the output features output by the decoder module immediately preceding the decoder module. Below decoder layer 1, an adaptive classification head 1 is connected in series. The adaptive classification head 1 classifies and predicts the output features of decoder layer 1 to obtain a first fault classification result, and updates the first fault classification result based on a preset classification calibration matrix to obtain a second fault classification result output by the decoder module. That is, the configurations of each decoder module are basically the same, and each obtains a first fault classification result and a second fault classification result.

[0029] When all the processing of the n decoder modules is completed, the second fault classification result output by the decoder module of the last layer of the decoder network is obtained.

Number

[0030]

Number

[0031] In an embodiment of the present disclosure, the decoder network is constructed based on a Transformer network structure.

[0032] In an embodiment of the present disclosure, the decoder network includes a plurality of decoder modules connected in series in sequence. Each decoder module includes a decoder layer and an adaptive classification head, and the adaptive classification head is used to classify and predict the output features of the decoder layer to obtain a first fault classification result. The method of using the adaptive classification head aims to calibrate the prediction results of each layer based on the adaptive classification head to improve the detection performance. In an embodiment of the present disclosure, image collection is performed on the back surface of the spinning box in the spinning workshop to obtain an image of the processing target, and further, a first image feature is extracted from the image of the processing target. The first image feature is processed based on the decoder network to obtain a fault detection result for the process control device. This method helps to automate the monitoring of the spinning box in the spinning workshop and timely discover the fault area and fault type.

[0033] The overall flow of the embodiment of the present disclosure can be divided into three parts: acquisition of the first image feature, optimization of the decoder network, and acquisition of the fault detection result. These three parts will be described in detail below.

[0034] 1) Acquisition of the first image feature

[0035] In some embodiments, extracting the first image feature from the image of the processing target can be implemented as follows.

[0036] In step A1, the image of the processing target is input into the backbone network to obtain initial image features.

[0037] Examples of the adoptable backbone network include, for example, vggnet (Visual Geometry Group Net, a computer vision network), resnet (residual network), and inception (startup) network. Of course, the embodiments of the present disclosure are not limited to a specific backbone network.

[0038] In step A2, the initial image features are input into the encoder network to obtain the first image features.

[0039] Here, the encoder network may be a Transformer encoder.

Number

[0040] In the embodiments of the present disclosure, the backbone network and the encoder network are used to perform feature extraction on the processed image, and the first image features for fault detection can be accurately obtained.

[0041] In some embodiments, as described above, since the temperature control of the heat preservation layer is important for the spinning process, in the embodiments of the present disclosure, when collecting the thermal image of the spinning workshop based on the infrared camera, extracting the first image features from the image to be processed can be further implemented as follows.

[0042] In step B1, the image to be processed is input into the segmenting and enhancing model to obtain the mask image of the target position.

[0043] Here, the image to be processed is input into the segmenting and enhancing model. As shown in FIG. 3, the image to be processed is input into the image feature extraction network to obtain the encoded features. At least one convolutional layer is used to perform convolutional processing on the preset mask to obtain the mask features. Here, the preset mask may be a mask based on points or frames, or a mask obtained by segmenting an object based on other segmentation networks. After the mask features and the encoded features are fused, they are input into the mask decoder. At the same time, the presentation information (including at least one of points, frames, and texts) is also input into the mask decoder. After being processed by the mask decoder, the mask image of the target position can be obtained. The target position in the embodiments of the present disclosure may be the damaged position in the heat preservation layer.

[0044] During implementation, image samples of the damaged positions of the heat preservation layer can be collected in advance. Based on the image samples, the characteristics of the damaged positions are constructed. For example, statistical analysis can be performed on the pixel values of the damaged positions, and the distribution of the pixel values can be obtained as the characteristics of the damaged positions. The distribution of the pixel values can be represented by the amount of pixels in different pixel value intervals. Further, for example, feature extraction can be performed on the damaged positions to obtain the characteristics of the damaged positions.

[0045] In the embodiments of the present disclosure, images are collected at fixed points on the specified inspection route by controlling a drone. That is, for the same detection area in the process control device of the spinning box, the contents of the continuously collected images are substantially similar. Thus, in the embodiments of the present disclosure, a plurality of candidate detection frames can be pre-marked on the images of the processing targets collected for the same detection area. For each image of the processing target, pixel points or pixel frames are selected from each candidate detection frame based on the characteristics of the damaged positions, and the presentation information of the segmentation and enhancement model is constructed.

[0046] Specifically, when the characteristic of the damaged position is the pixel value distribution, a plurality of pixel points are obtained within the m-neighborhood from any point in the candidate detection frame, and then the pixel distribution to be compared is statistically obtained. The pixel distribution to be compared is matched with the characteristic of the damaged position, and the matching degree is calculated. When the matching degree is higher than the threshold value, the points and / or frames in the presentation information of the segmentation and enhancement model are obtained. m is a positive integer.

[0047] When the characteristics of the damaged positions are obtained by feature extraction, feature extraction is performed on the candidate detection frames to obtain the characteristics to be compared. The characteristics to be compared are matched with the characteristics of the damaged positions. When the matching degree is higher than the threshold value, the points and / or frames in the presentation information of the segmentation and enhancement model are obtained.

[0048] In step B2, based on the mask image of the target position, a minimum rectangular bounding box is generated to enclose the mask image of the target position.

[0049] The points and frames in the pre-constructed presentation information can be understood to roughly determine the damage position. By using a segment enclosing model and improving and optimizing the damage position based on the presentation information, a more accurate mask image can be obtained.

[0050] In order to improve the accuracy of fault detection, in the embodiments of the present disclosure, the features of the target position are described from multiple aspects. Specifically, it can be shown as from step B3 to step B7.

[0051] In step B3, based on the minimum rectangular bounding box, a visible sub-image is cut out from the image to be processed, and a thermal sub-image in the minimum rectangular bounding box is cut out from a thermal image in the same view range as the image to be processed.

[0052] In step B4, the visible sub-image is input into a backbone network to obtain first sub-image features.

[0053] In step B5, the first sub-image features are input into an encoder network to obtain sub-image features to be fused.

[0054] In step B6, thermal features are extracted from the thermal sub-image.

[0055] It should be noted that as long as the execution timing of step B6 is reasonable, it is not limited to being executed after step B5.

[0056] Here, a long short-term memory (LSTM) network can be used to process thermal sub-images to obtain thermal features. For example, a drone can be controlled to follow the same inspection route and collect images at fixed points. The viewing ranges when obtaining the images are approximately the same. Therefore, in multiple inspections, the objects included in multiple images at the same position point are approximately the same. Obtain the content of the images in the minimum bounding box in multiple thermal images within a preset period to obtain a sequence of thermal sub-images, extract the temperature change trend features based on the LSTM, and use them as thermal features.

[0057] In step B7, fuse the sub-image features of the fusion target, the thermal features, and the second sub-image features extracted from the minimum bounding box of the image to be processed by the segmentation-engining model to obtain the first image features.

[0058] Here, the fusion method may be weighted fusion, concatenation fusion, etc.

[0059] In the embodiments of the present disclosure, a segmentation-engining model can be used to process the image to be processed to obtain a mask image of the target position, and further obtain a visible sub-image. Input the visible sub-image into a backbone network and an encoder network to obtain the sub-image features of the fusion target. Further, based on the visible sub-image, obtain the thermal sub-image of the visible sub-image, and then extract the thermal features of the thermal sub-image. Fuse the sub-image features of the fusion target, the thermal features, and the second sub-image features extracted from the minimum bounding box of the image to be processed by the segmentation-engining model to obtain the first image features. The first image features can be extracted from multiple dimensions based on the above method. In particular, the thermal features can describe the temperature change trend of the target position, laying a strong foundation for obtaining subsequent fault detection results.

[0060] 2) Optimization of the decoder network

[0061] In order to further improve the detection accuracy, based on FIG. 2, a query ordering layer can be added to optimize the decoder network. As shown in FIG. 4, decoder modules 1, 2, …, n are connected in series in sequence. Except for the last layer's decoder module, at least one other decoder module includes a decoder layer, an adaptive classification head, and a query ordering layer. An adaptive classification head 1 is connected in series under the decoder layer 1 in the decoder module, a query ordering layer 1 is connected in series under the adaptive classification head 1, and then the configurations in decoder module 2 are connected in series in sequence, and so on by analogy. Each decoder layer can include a self-attention module and a cross-attention module (not shown) connected after the self-attention module. When the decoder network includes n decoder modules, the query vector is iteratively adjusted n times.

Number

Number

[0062] The first sub-feature is used to capture semantic information, and the second sub-feature is used, for example, to encode the location information of a fault such as the center and size distribution of the bounding box.

[0063] Hereinafter, a method for updating two sub-features in the query vector will be described. Specifically, it includes the following.

[0064] 1) Update of the first sub-feature

[0065] As described above, the query vector of each decoder module includes a first sub-feature. At least one decoder module except the first decoder module in the decoder network further includes a query ordering layer. For the decoder module including the query ordering layer, the decoder module updates the first sub-feature of the decoder module based on the following method.

[0066] In step C1, based on the second fault classification result output by the decoder module, the first element in the first sub-feature of the decoder module is sorted in descending order to obtain a first intermediate feature.

[0067] Here, the first intermediate feature of the decoder module in the L-1 layer is shown, for example, in Equation (2).

Number

[0068] Each decoder module not only outputs a second fault classification result but also outputs w detection frames, where w is a positive integer. Accordingly, the first sub-feature includes w elements, and each element in the first sub-feature corresponds to one detection frame. p L-1 Similarly includes w elements, each element corresponding to the fault classification result of one detection frame, the preset number of fault types is m, and the fault classification result of any of the detection frames is a probability distribution {m1, m2,..., mm} for each preset fault type, where m1 represents the probability that the detection frame belongs to fault type 1 and m2 represents the probability that the detection frame belongs to fault type 2. Using the maximum probability in the fault classification result of the detection frame, the elements in the first sub-feature are sorted in descending order.

[0069] In step C2, the first intermediate feature is concatenated with a preset content calibration matrix to obtain a first concatenated feature.

[0070] In step C3, process the first stitching feature based on the fully connected layer to obtain the first sub-feature required for the next decoder module of the decoder module.

[0071] After obtaining the first intermediate feature of the decoder module in the L-1 layer, the first sub-feature of the next decoder module is shown, for example, in Equation (3).

Equation

[0072] In the embodiments of the present disclosure, the first intermediate feature of the previous layer is ordered using the second fault classification result of the previous layer, and the first sub-feature of the layer is adjusted based on the first intermediate feature and a preset content calibration matrix, so as to optimize the ordering, thereby obtaining the first sub-feature required by the decoder module of the layer, which is more accurately described.

[0073] As described above, the query vector further includes a second sub-feature for encoding the position information of the fault, and the number of elements of the first sub-feature and the second sub-feature is the same.

[0074] For the decoder module including the query ordering layer, in order to match the order of the second sub-feature with the ranked first sub-feature, the second sub-feature is ordered or reconstructed based on different DETR-based detectors for the second sub-feature.

[0075] In one possible embodiment, for the H-DETR detector (hybrid matching approach), by using the same second sub-feature for all layers of the decoder module, only the second sub-feature of the previous layer is ordered, and the decoder module updates the second sub-feature of the decoder module based on the following method.

[0076] The update method of the second sub-feature of the decoder module is shown, for example, in Equation (4).

Number

[0077] During implementation, taking w detection frames as an example, the second sub-feature of the decoder module in the L-1 layer contains w elements, and each element in the second sub-feature of the decoder module in the L-1 layer corresponds to one detection frame. p L-1 Similarly contains w elements, each element corresponding to the fault classification result of one detection frame, and the preset fault types are m. The fault classification result of any of the detection frames is a probability distribution {m1, m2,..., mm} for each preset fault type, where m1 represents the probability that the detection frame belongs to fault type 1, and m2 represents the probability that the detection frame belongs to fault type 2. Using the maximum probability in the fault classification result of the detection frame, the elements corresponding to each detection frame in the second sub-feature of the decoder module in the L-1 layer are sorted in descending order to obtain the second sub-feature of the decoder module in the L layer.

[0078] In another possible embodiment, for the DINO-DETR detector (DETR with Improved deNoising anchOr boxes), the decoder module updates the second sub-feature of the decoder module based on the following method.

[0079] In step D1, based on the second fault classification result output by the decoder module, the second sub-feature of the decoder module is sorted in descending order to obtain a second intermediate feature.

[0080] Here, the second intermediate feature of the decoder module in the L layer is shown, for example, in Equation (5).

Number

[0081] During implementation, the specific sorting method is similar to the sorting method of the first sub-feature, and the embodiments of the present disclosure will not repeat this here.

[0082] In step D2, the second intermediate feature is reconstructed based on the sine wave position encoding function and the multi-layer perceptron to obtain the second sub-feature required for the next decoder module of the decoder module.

[0083] The second sub-feature of the decoder module in the L-th layer is shown, for example, in Equation (6).

Equation

[0084]

Equation

[0085] In Equation (7), (u, v) represents the coordinate position of each point in the detection frame in the image to be processed, D represents the encoding dimension required for the second sub-feature, usually D is a predetermined value, T is the modulation temperature, and generally 10000 is desirable.

[0086] After the position encoding of Equation (7), the updated second sub-feature is obtained by multi-layer perceptron processing.

[0087] In the embodiments of the present disclosure, by reconstructing the second intermediate feature based on the sine wave position encoding function and the multi-layer perceptron and obtaining the second sub-feature required for the next decoder module, the detection frame of the fault object can be described more accurately, and subsequent processing can be facilitated.

[0088] In some embodiments, the use of the embodiments of the present disclosure in a previously trained decoder network has been described. In the embodiments of the present disclosure, so as to obtain an appropriate decoder network, in the training process of the decoder network, the classification calibration matrix of each decoder module in the decoder network is learnable and is obtained by training the decoder network. It can be implemented as follows.

[0089] In the process of training the decoder network, the control of the temperature of the melt in the spinning workshop is relatively important. The identification of faults focuses on evaluating the heat preservation effect of the heat preservation layer. Randomly initializing the classification calibration matrix may deviate from the ideal state. In order to accelerate the convergence of the model, in the embodiments of the present disclosure, the classification calibration matrix can be initialized based on the following method.

[0090] In step H1, a sub-sample of the fault area is obtained by dividing the samples of the thermal images collected for the process control device.

[0091] In step H2, the classification calibration vector is initialized based on the difference between the sub-sample and the sample data of the fault area in the normal operating state.

[0092] In the embodiments of the present disclosure, by initializing the classification calibration vector based on the difference between the sub-sample and the sample data of the fault area in the normal operating state, the initial state of the classification calibration vector can be represented based on the temperature information of the heat preservation layer, thereby facilitating the acceleration of the convergence of model training and assisting in learning the available classification calibration vector as early as possible.

[0093] 3) Obtaining the fault detection result

[0094] In some embodiments, as described above, the process control device includes a temperature control system, and the temperature control system includes a heating device and a melt distribution pipe. The temperature detection can be implemented as follows.

[0095] In step E1, based on an infrared camera, a thermal image of the temperature control system is obtained.

[0096] In step E2, for the target component of the temperature control system, an image block of the target component is extracted from the thermal image.

[0097] Here, the target component may be a melt distribution pipe or a heating device.

[0098] In step E3, by comparing the image block of the target component with the standard image block of the target component, difference information between the image block of the target component and the standard image block of the target component is obtained.

[0099] The standard image block of the target component is the standard temperature range of the target component. The collected image block of the target component is compared with the standard temperature range of the target image to obtain difference information between the two. Specifically, the difference information can be expressed based on statistical information of the difference in pixel values at the same pixel position. The statistical information is, for example, the average value, variance, etc. Also, using a pre-trained model, the feature matrices of the two are respectively extracted, and then the difference between the feature matrices is calculated to obtain the difference information.

[0100] In step E4, based on the difference information, it is detected whether the temperature of the target component is within the normal temperature range.

[0101] When the image block of the target component belongs to the standard temperature range of the target image, it is determined that there is no abnormality in the target component. On the other hand, when the image block of the target component does not belong to the standard temperature range of the target image, it is determined that there is an abnormality in the target component. The second fault classification result output by the decoder module in the last layer of the decoder network in the fault detection result and the fault position output by the decoder module in the last layer are combined to determine the abnormal situation.

[0102] In the embodiments of the present disclosure, the thermal image of the collected target component is compared with the standard image block of the target component to determine whether the temperature of the target component is normal at that time.

[0103] In some embodiments, the second fault classification result in the fault detection result includes a fault type and a fault grade, and the method further includes the following.

[0104] In step F1, when the fault detection result indicates that a fault has occurred in the temperature control system of the process control device and the fault level is higher than a preset level, a repair task is generated, and the repair task includes the urgency of the repair task and the fault detection result.

[0105] In step F2, the repair task is pushed to the target client side.

[0106] Here, for example, if it is detected that the damage to the heat preservation layer on the upper surface of the melt distribution pipe in the temperature control system is too large (for example, the depth of the damage is greater than a preset depth and the radius of the damage is greater than a preset radius), and it affects the temperature of the heat preservation pipe (for example, an abnormality is discovered based on a thermal image), the fault detection result is the highest risk and immediate processing is required. If it is detected that the damage to the heat preservation layer on the upper surface of the melt distribution pipe in the temperature control system is too small and does not affect the temperature of the heat preservation pipe, the fault detection result is a low risk, and the repair task is first recorded and can be processed regularly by the employee.

[0107] In the embodiments of the present disclosure, different processing strategies are executed based on the urgency of the fault and can be adaptively adjusted based on the actual situation.

[0108] In some embodiments, since there is a possibility that the auxiliary materials of the process may accumulate in the open area of the spinning workshop, it is necessary to collect images for the open area as well, and it can be implemented as follows.

[0109] In step G1, an image is collected for the open area of the spinning box in the spinning workshop to obtain a target image.

[0110] In step G2, based on the target image, the accumulation status of the auxiliary materials of the process in the open area is detected to obtain a detection result.

[0111] In step G3, a detection record is generated based on the detection result and the historical detection result of the open area.

[0112] In the case of a detection result indicating the presence of a deposition of auxiliary materials for the process at position coordinate 1, since there are already two records of "there is a deposition of auxiliary materials for the process at position coordinate 1" in the historical detection result, the detection result is continuously recorded as the third record of "there is a deposition of auxiliary materials for the process at position coordinate 1".

[0113] In step G4, the detection record is matched with the set of reporting strategies.

[0114] Here, the set of reporting strategies is such that when there is one same detection record, it is reported to a first-level employee; when there are two same detection records, it is reported to a second-level employee; when q same detection records are recorded, it is reported to a q-level employee, and so on by analogy. The q-level manager is the superior manager of the (q - 1)-level manager, and by analogy, the second-level manager is the superior of the first-level manager.

[0115] In step G5, warning information is generated based on the reporting strategy obtained by matching in the set of reporting strategies.

[0116] In the embodiments of the present disclosure, in order to complement management vulnerabilities, different reporting strategies are corresponding based on the deposition situation of the auxiliary materials for the process, thereby avoiding the operator forgetting.

[0117] To summarize the above, in one possible embodiment, the network framework applicable to the detection method of the spinning workshop proposed in the embodiments of the present disclosure is, for example, as shown in FIG. 5. The image to be processed is input into the backbone network to obtain initial image features, and the initial image features are input into the encoder network to obtain the first image features. The first image features are input into a decoder network including n decoder modules.

[0118] In an embodiment of the present disclosure, the query vector is initialized to Q 0 is obtained, the processed image is divided into a plurality of grids, each grid is position-encoded, and the target position encoding PE is obtained. The initialized query vector Q 0 and the target position encoding PE are fused and then used as the query vector Q of the self-attention module 1 in the first decoder module 1, and the target position encoding PE is used as the K and V of the self-attention module 1 in the first decoder module 1. The output of the self-attention module 1 is the self-attention feature TE1.

[0119] The self-attention feature TE1 and the target position encoding PE are fused and used as the Q of the cross-attention module 1 in the first decoder module 1, the first image feature is used as the V of the cross-attention module 1, and the fused feature of the first image feature and the preset position encoding CE in the input of the encoder network (the position encoding is the position encoding of the camera, for example, the position encoding in the camera's patrol route) is used as the K of the cross-attention module 1.

[0120] For each subsequent decoder module L, the Q, K, and V of its self-attention module L are constructed based on the target position encoding PE and the updated Q of the previous decoder module. For example, both K and V are the target position encoding PE, and its Q is the Q updated based on the previous decoder module.

[0121] For each subsequent decoder module L, the Q, K, and V of its cross-attention module L are constructed based on the self-attention feature TEL, the first image feature, and the preset position encoding. For example, its Q is the self-attention feature TEL output by the self-attention module L, its V is the first image feature, and K is the fused feature of the first image feature and the preset position encoding.

[0122] In the processing process, through the processing of each layer of decoder layer, adaptive classification head, and query ordering layer, the final fault detection result and fault location are obtained.

[0123] Based on the same technical concept, an embodiment of the present disclosure provides a detection device 600 for a spinning workshop as shown in FIG. 6, a collection unit 601 for collecting images of a process control device of a spinning box in a spinning workshop to obtain an image to be processed, an extraction unit 602 for extracting first image features from the image to be processed, and a processing unit 603 for processing the first image features based on a decoder network to obtain a fault detection result for the process control device. The decoder network includes a plurality of decoder modules connected in series in sequence, For each decoder module, the decoder module includes a decoder layer and an adaptive classification head, and the adaptive classification head is used to classify and predict the output features of the decoder layer to obtain a first fault classification result, and update the first fault classification result based on a preset classification calibration matrix to obtain a second fault classification result output by the decoder module. The fault detection result output by the decoder network includes a second fault classification result output by the decoder module of the last layer of the decoder network and a fault position output by the decoder module of the last layer.

[0124] In some embodiments, the query vector of each decoder module includes a first sub-feature for capturing semantic type information, At least one decoder module other than the last decoder module in the decoder network further includes a query ordering layer, The device is For the decoder module including the query ordering layer, by the decoder module, Based on the second fault classification result output by the decoder module, the first element in the first sub-feature of the decoder module is sorted in descending order to obtain a first intermediate feature, The first intermediate feature is combined with a preset content calibration matrix to obtain a first combined feature, A first update unit is further provided for updating the first sub-feature of the decoder module based on a method of processing the first stitching feature based on a fully connected layer and obtaining a first sub-feature required for the next decoder module of the decoder module.

[0125] In some embodiments, the query vector of each decoder module further includes a second sub-feature for encoding the location information of the fault, and the number of elements of the first sub-feature and the second sub-feature is the same. The apparatus For the decoder module including the query ordering layer, by the decoder module, Based on the second fault classification result output by the decoder module, the second sub-feature of the decoder module is sorted in descending order to obtain a second intermediate feature. A second update unit is further provided for updating the second sub-feature of the decoder module based on a method of reconstructing the second intermediate feature based on a sine wave position encoding function and a multi-layer perceptron and obtaining a second sub-feature required for the next decoder module of the decoder module.

[0126] In some embodiments, the second fault classification result in the fault detection result includes a fault type and a fault level, and the apparatus When the fault detection result indicates that a fault has occurred in the temperature control system of the process control device and the fault level is higher than a preset level, generating a repair task including the urgency of the repair task and the fault detection result; And further includes a push unit for pushing the repair task to the target client side.

[0127] In some embodiments, the extraction unit 602 Inputs the image to be processed into a backbone network to obtain initial image features; And inputs the initial image features into an encoder network to obtain the first image features.

[0128] In some embodiments, the extraction unit 602 inputs the image to be processed into a segmentation engine model to obtain a mask image of the target position, generates a minimum bounding rectangle based on the mask image of the target position so as to accommodate the mask image of the target position, cuts out a visible sub-image from the image to be processed based on the minimum bounding rectangle, and cuts out a thermal sub-image in the minimum bounding rectangle from a thermal image within the same view range as the image to be processed, inputs the visible sub-image into a backbone network to obtain first sub-image features, inputs the first sub-image features into an encoder network to obtain sub-image features to be fused, extracts thermal features from the thermal sub-image, and is used to perform a fusion process on the sub-image features to be fused, the thermal features, and second sub-image features extracted by the segmentation engine model from the minimum bounding rectangle of the image to be processed, to obtain the first image features.

[0129] In some embodiments, the process control device includes a temperature control system, and the device acquires a thermal image of the temperature control system based on an infrared camera, extracts an image block of the target component from the thermal image for the target component of the temperature control system, obtains difference information between the image block of the target component and the standard image block of the target component by comparing the image block of the target component with the standard image block of the target component, and further includes a temperature detection unit used to detect whether the temperature of the target component is within a normal temperature range based on the difference information.

[0130] In some embodiments, the device Collect images of the open area of the spinning box in the spinning workshop to obtain a target image, and Based on the target image, detect the deposition status of the auxiliary materials in the open area to obtain a detection result, and Based on the detection result and the historical detection result of the open area, generate a detection record, and Match the detection record with a set of report strategies, and Further include an alarm unit used to generate alarm information based on the report strategy obtained by matching in the set of report strategies.

[0131] In some embodiments, the classification calibration matrix of each decoder module in the decoder network is learnable and obtained by training the decoder network. The device In the process of training the decoder network, Obtain a sub-sample of the fault area by dividing a sub-sample of the thermal image collected for the process control device, Further include an initialization unit used to initialize the classification calibration matrix based on a method of initializing the classification calibration vector based on the difference between the sub-sample and the sample data of the fault area in the normal operating state.

[0132] For the specific functions and exemplary descriptions of each module, sub-module / unit of the device according to the embodiments of the present disclosure, reference can be made to the relevant descriptions of the corresponding steps in the above-described method embodiments, which will not be repeated here.

[0133] In the technical solution of the present disclosure, the acquisition, storage, and application of the user's personal information comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0134] FIG. 7 is a structural block diagram of an electronic device according to an embodiment of the present disclosure. As shown in FIG. 7, the electronic device includes a memory 710 and a processor 720, and a computer program executable by the processor 720 is stored in the memory 710. The number of the memory 710 and the processor 720 can be one or more. The memory 710 can store one or more computer programs, and when the one or more computer programs are executed by the electronic device, the electronic device is caused to execute the method provided by the above method embodiment. The electronic device can further include the following. A communication interface 730 is used to communicate with an external device and perform data interaction and transmission.

[0135] When the memory 710, the processor 720, and the communication interface 730 are independently implemented, the memory 710, the processor 720, and the communication interface 730 are connected to each other via a bus and can communicate with each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be classified into an address bus, a data bus, a control bus, etc. For ease of explanation, only a single thick line is shown in FIG. 7, but it does not indicate only a single bus or one type of bus.

[0136] Optionally, in a specific implementation form, when the memory 710, the processor 720, and the communication interface 730 are integrated on one chip, the memory 710, the processor 720, and the communication interface 730 can communicate with each other via an internal interface.

[0137] The above-mentioned processor may be a Central Processing Unit (CPU), and it should be understood that it may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware assemblies, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. Note that the processor may be a processor that supports the Advanced RISC Machines (ARM) architecture.

[0138] Furthermore, optionally, the memory may include a read-only memory and a random access memory, and may further include a non-volatile random access memory. The memory can be either a volatile memory or a non-volatile memory, or can include both a volatile memory and a non-volatile memory. Here, the non-volatile memory can include ROM (Read-Only Memory), PROM (Programmable ROM), EPROM (Erasable PROM), EEPROM (Electrically EPROM), or flash memory. The volatile memory can include a random access memory (Random Access Memory, RAM) that functions as an external cache. By way of example and not limitation, many forms of RAM are available. For example, Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct RAMBUS RAM (DR RAM).

[0139] In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, all or part of it may be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the processes or functions according to the embodiments of the present disclosure are generated in whole or in part. The computer may be a general-purpose computer, a dedicated computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (such as infrared, Bluetooth®, microwave, etc.). The computer-readable storage medium may be any available medium accessible by a computer, or a data storage device including a server, data center, etc. integrated with one or more available media. The available medium may be a magnetic medium (such as a floppy (registered trademark) disk, hard disk, magnetic tape), an optical medium (such as a Digital Versatile Disc (DVD)), or a semiconductor medium (such as a Solid State Disk (SSD)), etc. It should be noted that the computer-readable storage medium mentioned in the present disclosure may be a non-volatile storage medium, in other words, a non-transitory storage medium.

[0140] Those skilled in the art can understand that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by instructing relevant hardware through a program. The program may be stored in a computer-readable storage medium, and the above storage medium may be a read-only memory, a magnetic disk, an optical disk, or the like.

[0141] In the description of the embodiments of the present disclosure, the description of reference terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or features described in relation to the embodiment or example are included in at least one embodiment or example of the present disclosure. And the specific features, structures, materials, or features described can be combined in any one or more embodiments or examples in an appropriate manner. Furthermore, those skilled in the art may combine different embodiments or examples described in the present disclosure and the features of different embodiments or examples without contradiction.

[0142] In the description of the embodiments of the present disclosure, " / " represents "or" unless otherwise described. For example, A / B may represent either A or B. "And / or" in the present disclosure only describes the relationship of related objects and indicates that three types of relationships may exist. For example, A and / or B can indicate the following. There are three situations where A exists alone, A and B exist simultaneously, and B exists alone.

[0143] In the description of the embodiments of the present disclosure, the terms "first" and "second" are used only for the purpose of description and should not be construed as indicating or implying relative importance or indicating the number of the technical features shown. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present disclosure, "a plurality" means two or more unless otherwise described.

[0144] The above are only exemplary embodiments of the present disclosure and do not limit the present disclosure. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle scope of the present disclosure should all be included within the protection scope of the present disclosure.

Claims

1. A method for detecting a spinning workplace, comprising: collecting an image of a process control device of a spinning box in a spinning workplace to obtain an image to be processed; extracting a first image feature from the image to be processed; processing the first image feature based on a decoder network to obtain a fault detection result for the process control device; the decoder network includes a plurality of decoder modules connected in series in sequence; for each decoder module, the decoder module includes a decoder layer and an adaptive classification head, and the adaptive classification head is used to classify and predict the output features of the decoder layer to obtain a first fault classification result, and update the first fault classification result based on a preset classification calibration matrix to obtain a second fault classification result output by the decoder module; the fault detection result output by the decoder network includes a second fault classification result output by the decoder module of the last layer of the decoder network and a fault position output by the decoder module of the last layer; for each decoder module after the first decoder module in the decoder network, taking the output features of the decoder module immediately preceding the decoder module as the input features of the decoder module; A method for detecting a spinning workplace.

2. The query vector of each decoder module includes a first sub-feature for capturing semantic type information; at least one decoder module other than the last decoder module in the decoder network further includes a query ordering layer; for the decoder module including the query ordering layer, the decoder module: orders the first element in the first sub-feature of the decoder module in descending order based on the second fault classification result output by the decoder module to obtain a first intermediate feature; combines the first intermediate feature with a preset content calibration matrix to obtain a first combined feature; processes the first combined feature based on a fully connected layer to obtain a first sub-feature required for the next decoder module of the decoder module, and updates the first sub-feature of the decoder module based on the method; The method for detecting a spinning workplace according to Claim 1.

3. The query vector of each decoder module further includes a second sub-feature for encoding the position information of the fault, and the number of elements of the first sub-feature and the second sub-feature is the same. Regarding the decoder module including the query ordering layer, Based on the second fault classification result output by the decoder module, the second sub-feature of the decoder module is sorted in descending order to obtain a second intermediate feature. Reconstruct the second intermediate feature based on the sine wave position encoding function and the multi-layer perceptron, and update the second sub-feature of the decoder module based on the method for obtaining the second sub-feature required for the next decoder module of the decoder module. The detection method for a spinning mill according to claim 2.

4. The second fault classification result in the fault detection result includes the fault type and the fault level. The detection method for the spinning mill is as follows. When the fault detection result indicates that a fault has occurred in the temperature control system of the process control device and the fault level is higher than a preset level, it is a repair task, and a repair task including the urgency of the repair task and the fault detection result is generated. Pushing the repair task to the target client side. The detection method for a spinning mill according to claim 1.

5. Extracting the first image feature from the image to be processed includes: Inputting the image to be processed into the backbone network to obtain an initial image feature. Inputting the initial image feature into the encoder network to obtain the first image feature. The detection method for a spinning mill according to claim 1.

6. Extracting the first image feature from the image to be processed includes: Inputting the image to be processed into the segmenting engine model to obtain a mask image of the target position. Based on the mask image of the target position, generating a minimum rectangular bounding box that encloses the mask image of the target position. Cutting out a visible sub-image from the image to be processed based on the minimum rectangular bounding box, and cutting out a thermal sub-image in the minimum rectangular bounding box from a thermal image within the same view range as the image to be processed. Inputting the visible sub-image into the backbone network to obtain a first sub-image feature. Inputting the first sub-image feature into the encoder network to obtain a sub-image feature to be fused. Extracting a thermal feature from the thermal sub-image. Performing a fusion process on the sub-image features of the object to be fused, the thermal features, and the second sub-image features extracted by the segment-ensining model from the minimum bounding rectangle of the image to be processed to obtain the first image features, The method for detecting a spinning workplace according to claim 1.

7. The process control device includes a temperature control system, The method for detecting a spinning workplace includes: Obtaining a thermal image of the temperature control system based on an infrared camera; Extracting an image block of the target component from the thermal image for the target component of the temperature control system; Obtaining difference information between the image block of the target component and the standard image block of the target component by comparing the image block of the target component with the standard image block of the target component; Further including detecting whether the temperature of the target component is within a normal temperature range based on the difference information. The method for detecting a spinning workplace according to claim 1.

8. The method for detecting a spinning workplace includes: Collecting an image of the open area of the spinning box in the spinning workplace to obtain a target image; Detecting the deposition status of the auxiliary materials of the process in the open area based on the target image to obtain a detection result; Generating a detection record based on the detection result and the historical detection result of the open area; Matching the detection record with a set of report strategies; Further including generating alarm information based on the report strategy obtained by matching in the set of report strategies. The method for detecting a spinning workplace according to claim 1.

9. A detection device for a spinning workplace, comprising: A collection unit for collecting an image of the process control device of the spinning box in the spinning workplace to obtain an image to be processed; An extraction unit for extracting first image features from the image to be processed; A processing unit for processing the first image features based on a decoder network to obtain a fault detection result for the process control device, The decoder network includes a plurality of decoder modules connected in series in sequence. For each decoder module, the decoder module includes a decoder layer and an adaptive classification head, and the adaptive classification head is used to classify and predict the output features of the decoder layer to obtain a first fault classification result, and update the first fault classification result based on a preset classification calibration matrix to obtain a second fault classification result output by the decoder module. The fault detection result output by the decoder network includes a second fault classification result output by the decoder module of the last layer of the decoder network and a fault position output by the decoder module of the last layer. For each decoder module after the first decoder module in the decoder network, the output features of the decoder module immediately preceding the decoder module are used as the input features of the decoder module. A detection device for a spinning workshop.

10. The query vector of each decoder module includes a first sub-feature for capturing semantic type information. At least one decoder module other than the last decoder module in the decoder network further includes a query ordering layer. The detection device for the spinning workshop, for the decoder module including the query ordering layer, by the decoder module, Based on the second fault classification result output by the decoder module, the first element in the first sub-feature of the decoder module is sorted in descending order to obtain a first intermediate feature. The first intermediate feature and a preset content calibration matrix are concatenated to obtain a first concatenated feature. Based on a fully connected layer, the first concatenated feature is processed to obtain a first sub-feature required for the next decoder module of the decoder module, and further includes a first update unit for updating the first sub-feature of the decoder module based on a method. The detection device for a spinning workshop according to claim 9.

11. The query vector of each decoder module further includes a second sub-feature for encoding the position information of the fault, and the number of elements of the first sub-feature and the second sub-feature is the same. The detection device for the spinning workshop, for the decoder module including the query ordering layer, by the decoder module, Based on the second fault classification result output by the decoder module, the second sub-feature of the decoder module is sorted in descending order to obtain a second intermediate feature. A second update unit is further provided for updating the second sub-features of the decoder module based on a method of reconstructing second intermediate features based on a sine wave position symbolization function and a multi-layer perceptron and obtaining second sub-features required for the next decoder module of the decoder module. The detection device for a spinning workplace according to claim 10.

12. The second fault classification result in the fault detection result includes a fault type and a fault level. The detection device for the spinning workplace When the fault detection result indicates that a fault has occurred in the temperature control system of the process control device and the fault level is higher than a preset level, a repair task, generating a repair task including the urgency of the repair task and the fault detection result. And a push unit used for pushing the repair task to the target client side. The detection device for a spinning workplace according to claim 9.

13. The extraction unit Inputs the image to be processed into a backbone network to obtain initial image features. Inputs the initial image features into an encoder network to obtain the first image features. The detection device for a spinning workplace according to claim 9.

14. The extraction unit Inputs the image to be processed into a segmenting engine model to obtain a mask image of the target position. Generates a minimum rectangular bounding box so as to accommodate the mask image of the target position based on the mask image of the target position. Cuts out a visible sub-image from the image to be processed based on the minimum rectangular bounding box, and cuts out a thermal sub-image in the minimum rectangular bounding box from a thermal image in the same view range as the image to be processed. Inputs the visible sub-image into a backbone network to obtain first sub-image features. Inputs the first sub-image features into an encoder network to obtain sub-image features to be fused. Extracts thermal features from the thermal sub-image. Fuses the sub-image features to be fused, the thermal features, and second sub-image features extracted by the segmenting engine model from the minimum rectangular bounding box of the image to be processed to obtain the first image features. The detection device for a spinning workplace according to claim 9.

15. The process control device includes a temperature control system. The detection device for the spinning workplace Obtaining a thermal image of the temperature control system based on an infrared camera; Extracting an image block of the target component from the thermal image for the target component of the temperature control system; Obtaining difference information between the image block of the target component and the standard image block of the target component by comparing the image block of the target component with the standard image block of the target component; Further comprising a temperature detection unit used for detecting whether the temperature of the target component is within a normal temperature range based on the difference information; The detection device for a spinning workplace according to claim 9.

16. The detection device for the spinning workplace is configured to: Collect an image for the open area of the spinning box in the spinning workplace to obtain a target image; Detect the deposition status of the auxiliary material of the process in the open area based on the target image to obtain a detection result; Generate a detection record based on the detection result and the historical detection result of the open area; Match the detection record with a set of reporting strategies; Further comprising an alarm unit used for generating alarm information based on the reporting strategy obtained by matching in the set of reporting strategies; The detection device for a spinning workplace according to claim 9.

17. At least one processor; A memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is caused to execute the method according to any one of claims 1 to 8. An electronic device.

18. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the method according to any one of claims 1 to 8.

Images